Tencon 2005 – 2005 IEEE Region 10

Incorporating

ATNAC 2005

Supported by IEEE Victorian Section

Melbourne, Victoria, Australia

21 - 24 November 2005
ISBN 85590 814 9

Tencon 2005 – 2005 IEEE Region 10

© 2005 IEEE Personal use of this material is permitted. However permission to reprint/republish
the material for advertising or promotional purposes or for creating new collective works for resale
or redistribution to servers or lists, or to reuse any copyrighted component of this work in other
works must be obtained from IEEE.

IEEE Operations Center

445 Hoes Lane
Piscataway, NJ 08854-4150
USA

+1 800 678 IEEE (+1 800 678 4333)

+1 732 981 1393
+1 732 981 9667 (FAX)
email: customer-service@ieee.org

Published by Swinburne Press

Tencon 2005 Page 2

INDEX

Index .................................................................................................................................................................. 3

General Chair's Welcome.................................................................................................................................. 5

Tencon 05 TPC Chair Welcome........................................................................................................................ 5

IEEE Chair Welcome ......................................................................................................................................... 6

Registration and Social Program....................................................................................................................... 6

Tencon 05 Committees...................................................................................................................................... 7

About the IEEE .................................................................................................................................................. 9

Benefits of IEEE Membership:........................................................................................................................... 9

Conference Centre Map ...................................................................................................................................11

Tencon Tutorial Sessions – 21st November 2005 ........................................................................................... 13

Tencon Technical Sessions – 22nd November 2005 ....................................................................................... 15

Tencon Technical Sessions – 23rd November 2005........................................................................................ 16

Tencon Technical Sessions – 24th November 2005........................................................................................ 17

Reviewers for Tencon 05................................................................................................................................. 18

Guest Speakers ............................................................................................................................................... 23

Mythology and Folklore of Network Protocols................................................................................. 23

Towards an Optical Internet ............................................................................................................ 24
Signal Processing In Telecommunications ..................................................................................... 25
Photonic Transport in Biological Tissue - Application of Optical Communications Techniques to
Bio-Sensing ..................................................................................................................................... 26
Triple Play: Converge on the Screen, not on the Bill ...................................................................... 27

Abstracts...........................................................................................................................................................29

Authors List.....................................................................................................................................................141

Tencon 2005 Page 3

Tencon 2005 Page 4
General Chair's Welcome
On behalf of the Tencon '05 Organising Committee, I would like to welcome you
to Tencon '05, held in the Crown Promenade Hotel, Melbourne from Monday 21
November to Thursday 24 November 2005.

As you will see from the enclosed program, we have 3 days full of papers from
leading researchers from around the Region and around the world. Over 500
papers will be presented in 102 sessions. These papers were selected by the
Technical Program Committee as representing the leading thinking on a variety
of important topics.

In addition, for those wanting to catch up with the latest advances in the current topical
technologies, eight half day tutorials will be held on Monday 21 November, delivered by leading
experts in their fields.

We hope you enjoy your stay in beautiful Melbourne.

Hugh Bradlow
Chair, Organising Committee, Tencon '05

TPC Chair Welcome

On behalf of the Technical Program Committee, I would like to welcome you to

Tencon 05. We have worked very hard to ensure that this conference will be able
to provide you with a good way of promoting your research and interacting with
your colleagues. The theme of our conference is “A Meeting Place for Converging
Technologies and People”, so we do hope that you take advantage of the
opportunity to meet and network with people working in your research field and
related areas. This year’s Tencon has also incorporated ATNAC 2005 and we are
delighted to have been able to include multiple streams and guest speakers to
address the many exciting issues emerging from the area of telecommunication
networks and applications.
We received over 800 submissions to our conference and accepted approximately 530 papers for
oral and poster presentation. Most papers represented collaborations between a number of
authors and, for the accepted papers in the conference, there are over 1300 individuals involved -
from approximately 40 different countries in the IEEE Region 10 and beyond. In addition to the
large number of papers from Australia, substantial contributions were also made by authors
coming from China, India, Japan, Thailand, Taiwan, Malaysia and Korea.
I would like to take this opportunity to thank the many people who worked tirelessly in the
background to ensure that this conference will be a success. In particular, we had nearly 200
reviewers helping us to review the papers and we had to process over 2000 individual reviews. To
my Technical Program Committee, I would like to thank them very much for their contribution to the
success of this program; I know that many people worked late into the night after their normal
daytime jobs had been completed in order to perform the many tasks required of them. To my
Deputy TPC Chairman, Dr Zahir Hussain, I am most grateful for your continued support and
assistance throughout the many difficult and busy times.
Finally, I hope that all attendees enjoy their stay in the World’s Most Liveable City – Melbourne –
and I trust that you will take away pleasant memories of this conference.

Richard Harris
TPC Chair, Tencon 05

Tencon 2005 Page 5

IEEE Victorian Section Chair Welcome
Dear Tencon '05 Colleague

Engineering is the best of all professions. It seems each week brings challenges which require a
quota of creative solutions. In a dynamic investment environment some of these solutions allow
new business plans to be developed justifying either the provision of services where none existed
before, or higher capability services or services to more people at an affordable price. So, both
from a technological and a business viewpoint, no other profession seems to advance on such a
broad front at such a quick pace. This is reflected in the papers being presented at this
conference. From power engineering to telecommunications, and biomedical engineering to
robotics, this is truly an interesting mix. I would like to welcome you to Tencon '05 and encourage
you to enjoy as broad a range of presentations as possible because interesting new ideas can
come from working across more than one discipline.

It goes almost without saying that the IEEE Victorian Section is proud of the work carried out by the
General Committee, chaired by Dr Hugh Bradlow and the Technical Program Committee headed
by Professor Richard Harris. I would like to acknowledge as well the foresight of my predecessor,
Peter Wills in bidding to bring this conference to Melbourne.

Enn Vinnal
Chair, IEEE Victorian Section
Melbourne, Australia

Registration and Social Program

Tencon '05 registration will be open from 8:30 on Monday, 21 November, and will remain open
during the welcome reception that evening. The registration desk will then open at 8 am on the
subsequent two days of the conference.

A Welcome Reception will be held from 6 - 8 pm on Monday, 21 November, in the pre-function

area of the Crown Promenade hotel. The Reception provides an opportunity to meet fellow
delegates and pick up the Conference satchel. All registered delegates are invited to attend. The
cost of this function is included in the registration fee. Companions are welcome to attend on
payment of an Accompanying Persons Welcome Reception Ticket ($50).

The Conference Dinner will be held on the evening of Wednesday, 23 November at the Crown
Palladium. Pre-dinner drinks commence at 6:30 pm, with dinner starting at 7 pm. The cost of this
function is also included in the registration fee for all delegate classes except students. Students
and companions are welcome to attend on payment of an Accompanying Persons Dinner Ticket
($110). The Crown Palladium is connected to the conference venue by aerobridge - delegates will
be directed there at the close of sessions on Wednesday.

In addition, we have a selection of pre and post conference tours available for delegates who wish
to see more of our beautiful city. For further information please contact the registration desk. Your
satchel also contains a brochure which lists some of the attractions of Melbourne.

Tencon 2005 Page 6

Tencon 05 Committees
Local Committee
Position Name Affiliation
General Chair Professor Hugh Bradlow Telstra
Vice Chair Professor Rao Kotagiri University of Melbourne
Technical Program Committee Chair Professor Richard Harris Massey University, New Zealand
Vice Chair TPC & Publications Chair Dr Zahir Hussain RMIT University
Secretary Mr Roger Venning Telstra
Treasurer Mr Anthony Gascoigne GNS Associates
Publicity & Publications Dr Horace King Swinburne University of Technology
Marketing and Promotion Mr Craig Skinner Ovum Australia
Venue & Monash University CMO Liaison Ms Catherine Edwards Telstra
International Advisor Board Co-ordinator Professor Laurence Cahill La Trobe University
Web Co-ordinators Ms Sarah Craze ATCRC
Dr Charlotte Marra Operational Concepts
Mr Rhys Arkins Acme Packet
Exhibition Chair Dr Alexsander Kist RMIT University
Exhibition Advisor Richard Roxas Telstra
Publications Liaison Dr Irena Atov Swinburne University of Technology
Members at Large Dr Leith Campbell ATCRC
Mr Peter Wills Telstra
Dr Malin Premaratne Monash University
Monash CMO Ms Irene Thavarajah Monash University
Ms Helen Mc Lean Monash University
Ms Olga Lipkin Monash University
IEEE Victorian Section Representative Mr Enn Vinnal IEEE Victorian Section

Local Organising Committee TPC Members

Title Name Affiliation Country
Professor Richard Harris Massey University New Zealand
Dr Irena Atov Swinburne University of Technology Australia
Dr Alexander Kist RMIT University Australia
Dr Horace King Swinburne University of Technology Australia
Dr Zahir Hussain RMIT University Australia
Professor Laurie Cahill Latrobe University Australia
Dr Lachlan Andrew Melbourne University Australia
Dr Malin Premaratne Monash University Australia
A/Professor Grenville Armitage Swinburne University of Technology Australia
Professor Gerard Ledwich Queensland University of Technology Australia
Dr Peter O'Shea Queensland University of Technology Australia
Professor Saleh Al-Araji Etisalat College of Engineering UAE
Professor Aladin Zayegh Victoria University Australia
Professor Krzysztof Pawlikowski Canterbury University New Zealand
Professor Greg Egan Monash University Australia

International Advisory Committee Members

Title Name Affiliation Country
Professor Saleh Al-Araji Etisalat College UAE
Professor Yianni Attikiouzel Murdoch Australia
Professor Sanjay Bose NTU Singapore
A/Professor Paul Cheung University of Hong Kong Hong Kong
Dr P.K. Chiu HKIVE Hong Kong
Professor Trevor Cole University of Sydney Australia
Professor Serge Demidenko Monash University Malaysia Malaysia
Dr Chris Dick Xilinx USA
Professor Graham Goodwin Newcastle Australia
Professor Ray Jarvis Monash Australia
Professor Marzuki Khalid Universiti Teknologi Malaysia Malaysia
Professor P. Khincha IIT, Bangalore India
A/Professor Jae-Hyun Kim AJOU,Korea Korea
Dr Andrew Lee NTU Singapore
Dr Teck Seng Low Republic Polytechnic Singapore
Professor Jian Ma Nokia Research Research Centre (Beijing) China
A/Professor S.N. Merchant Indian Institute of Technology, Powai India
Dr Dalma Novak Pharad USA
Professor Kouichi Sakurai Hakozaki Japan
Mr Roberto Saracco Telecom Italia Italy
Dr S. Srikanth Anna University India
Professor Nguyen Tran IRIS Australia
A/Professor Laurence Wong Technical University of Singapore Singapore
Dr Sam Yu Ajou University Korea
Dr Bartek Wydrowski Caltech USA
Dr Dalibor Vrsalovic Intel USA
Dr Vasil Huatsysiu Rowan University USA

Tencon 2005 Page 7

Tencon 2005 Page 8
About the IEEE
The IEEE (Eye-triple-E) is a non-profit, technical professional association of more than 365,000
individual members in approximately 150 countries. The full name is the Institute of Electrical and
Electronics Engineers, Inc., although the organization is most popularly known and referred to by
the letters I-E-E-E.

Through its members, the IEEE is a leading authority in technical areas ranging from computer
engineering, biomedical technology and telecommunications, to electric power, aerospace
engineering and consumer electronics, among others.

Through its technical publishing, conferences and consensus-based standards activities, the IEEE
ƒ produces 30 percent of the world's published literature in electrical and electronics
engineering, and computer science areas,
ƒ holds annually more than 300 major conferences and
ƒ has nearly 900 active standards with almost 500 under development

The benefits of IEEE membership include these offerings:

ƒ Membership in one or more of 39 IEEE Societies and three Technical Councils spanning
the range of electrotechnologies, information technologies and more
ƒ More than 300 local organizations worldwide for member networking and information
sharing
ƒ Educational opportunities to ensure engineers' technical vitality
ƒ More than 1,300 student branches at universities in 80 countries
ƒ Special cost-saving and value-added benefits for MEMBERS ONLY
ƒ Prestigious awards and recognition of technical and professional achievements
ƒ Opportunities for volunteering, leadership and participation in a variety of IEEE activities
ƒ Electronic commerce with the IEEE through a variety of Web-based Services

IEEE Vision and Mission

Vision
To advance global prosperity by fostering technological innovation, enabling members' careers and
promoting community world-wide.

Mission
The IEEE promotes the engineering process of creating, developing, integrating, sharing, and
applying knowledge about electro and information technologies and sciences for the benefit of
humanity and the profession.

About the IEEE and Victorian Section

The Institute of Electrical and Electronics Engineers (IEEE) was founded in 1884 and the
Victorian Section was established nearly exactly on hundred years later in toward the end of
1983. The Section has previous been an active contributor to IEEE Region 10, hosting Tencon in
1992. The Section represents the Australian states of Victoria and Tasmania so that all IEEE
members who reside in Victoria or Tasmania are automatically members of our Section. The
Victorian Section has over 1,600 members. The Section is run by a committee of locally elected
IEEE members and includes the special interest groups called Chapters and Affinity Groups.Each
chapter or group has a Chair and is responsible for organising a number of interesting seminars,
site visits, workshops or conferences, throughout the year. The fifteen Chapters and Groups cater
for a wide range of technical interests and also areas like Women-in-Engineering, Education,
Graduates-of-the-Last-Decade. The Sectionhas three student branches.

Tencon 2005 Page 9

Tencon 2005 Page 10
Tencon 2005 Page 11
Tencon 2005 Page 12
 Tencon Tutorial Sessions – 21st November 2005

Session 1 Session 2 Session 3 Session 4

9:00 Network Internet Routing: Image Processing
Security Measurement & Modelling with FPGA
M2 M1 M3
Dr Radia Dr Matthew Dr Donald
Perlman Roughan Bailey
10:20 Morning Break
10:35 Network Internet Routing: Image Processing
Security Measurement & Modelling with FPGA
M2 M1 M3
Dr Radia Dr Matthew Dr Donald
Perlman Roughan Bailey
12:30 Lunch Break
13:15 MPLS and OmNet++ DSP for OFDM/ Digital
GMPLS Simulation Telecomms Communications
M2 M1 M3 M4
Alton Lo and Dr Ahmet Prof fred Dr Jean
Mallik Tatipamula Sekercioglu harris Armstrong
15:40 Afternoon Break
15:55 MPLS and OmNet++ DSP for OFDM/ Digital
GMPLS Simulation Telecomms Communications
M2 M1 M3 M4
Alton Lo and Dr Ahmet Prof fred Dr Jean
Mallik Tatipamula Sekercioglu harris Armstrong
18:00 Sessions conclude
Please note that lunch is not provided by the
conference and is not included in the tutorial fee

Tencon 2005 Page 13

Tencon 2005 Page 14
 Tencon Technical Sessions – 22nd November 2005
Promenade Room
8:40 Welcome by IEEE Victorian Section Chair – Mr Enn Vinnal
8:50 Welcome from Conference General Chair Dr Hugh Bradlow
9:00 Welcome by IEEE Region 10 Director - Professor Seiichi Takeuchi
9:05 Opening Speech: Mr Randall Straw, Executive Director, Multimedia Victoria
9:10 Plenary Session Chair: Dr Dalibor (Dado) F Vrsalovic
9:15 Speaker #1: Dr Radia Perlman
10:05 Speaker #2: Mr Mike Schwartz

ROOM 8:40 - 10:55 11:20 - 13:00 14:00 - 15:40 16:10 - 17:50

1B-01 1C-01
1D-01
P2 Traffic & Resource Traffic & Resource
Security Issues 1
Management 1 Management 2

1B-02 1C-02 1D-02

M1
Adhoc Networks Internet Applications 1 Internet Applications 2

1B-03 1C-03 1D-03

P1
Image Processing 1 Image Processing 2 Robotics

1B-04 1C-04 1D-04

P3
IT Applications 1 IT Applications 2 Microwave Circuits

1D-05
1B-05 1C-05
PLENARY SESSION REFER ABOVE

M2 Speech Signal
Modulation & Coding 1 Modulation & Coding 2
Processing & Coding

AFTERNOON BREAK
MORNING BREAK

1B-06 1C-06
LUNCH BREAK

1D-06
M3 Time-Frequency Analysis Time-Frequency Analysis
Power Transformers
and Wavelets 1 and Wavelets 2

1B-07 1C-07 1D-07

M4
Electronic Devices 1 Electronic Devices 2 Electronic Devices 3

1D-08
1B-08 1C-08
M6 Image Compression &
Biomedical Engineering 1 Biomedical Engineering 2
Quality Assessment

1B-09 1C-09 1D-09

M8 Code Division Multiple Code Division Multiple Estimation and
Access 1 Access 2 Modelling 1

1B-10 1C-10
1D-10
M10 Power Systems Planning Power Systems Planning
Informatics and AI
and Control 1 and Control 2

1B-11 1C-11 1D-11

M7
1 Future Challenges Miscellaneous Modelling Conversion Issues

1B-12 1C-12
1D-12
M9 Array and Space-Time Array and Space-Time
Power Generation
Signal Processing 1 Signal Processing 2

Tencon 2005 Page 15

 Tencon Technical Sessions – 23rd November 2005
Promenade Room
9:00 Speaker #3: Professor Rod Tucker
9:40 Speaker #4: Professor fred harris

ROOM 9:00 - 10:20 10:50 - 12:30 14:00 - 15:40 16:10 - 17:50

2B-01 2C-01 2D-01

M9 Network Performance and Network Performance and Internet Traffic
Design 1 Design 2 Management

2B-02 2C-02 2D-02

M7
Optical Networks Multimedia 1 Multimedia 2

2B-03 2C-03 2D-03

P2
Routing and Mobile IP Wireless LANS 1 Wireless LANS 2

2B-04 2C-04 2D-04

P1
Image Processing 3 Image Processing 4 Modulation & Coding 3

2D-05
2B-05 2C-05
PLENARY SESSION REFER ABOVE

P3 Time-Frequency Analysis
IT Applications 3 IT Applications 4
and Wavelets 3

AFTERNOON BREAK
MORNING BREAK

2B-06 2C-06 2D-06

LUNCH BREAK

M3 Power Quality Power Quality Signal Processing for

Monitoring 1 Monitoring 2 Media Applications

2B-07 2C-07 2D-07

M6 Adaptive Systems & Adaptive Systems & Adaptive Systems &
Signal Processing 1 Signal Processing 2 Signal Processing 3

2B-08 2C-08 2D-08

M4 Signal Processing Signal Processing Medical Engineering and
Hardware 1 Hardware 2 Agents

2B-09 2C-09 2D-09

M8 Othogonal Frequency Othogonal Frequency Estimation and
Division Multiplex 1 Division Multiplex 2 Modelling 2

2B-10 2C-10 2D-10

M10 Fuzzy Technique Fuzzy Technique IT Theory and
Applications 1 Applications 2 Applications

2B-11 2C-11 2D-11

M1
ATNAC Poster Session 1 ATNAC Poster Session 2 Tencon Poster Session 1

2B-12 2C-12 2D-12

M2
Tencon Poster Session 2 Tencon Poster Session 3 Tencon Poster Session 4

Tencon 2005 Page 16

 Tencon Technical Sessions – 24th November 2005
Promenade Room
8:00 Speaker #5: Professor Ray Jarvis
8:40 Speaker #6: Dr Malin Premaratne

ROOM 8:00 - 9:20 9:20 - 10:50 11:20 - 13:00 14:00 - 15:40

3B-01
3C-01 3D-01
M7 Mobile and Wireless
Convergence Issues Mobile Cellular
Networks

3B-02 3C-02 3D-02

P2
Computer Networks Security Issues 2 Security Issues 3

3B-03 3C-03 3D-03

P1 Filter Design and Filter Design and Filter Design and
Implementation 1 Implementation 2 Implementation 3

3B-04 3C-04 3D-04

PLENARY SESSION REFER ABOVE

P3 Robust and Adaptive Robust and Adaptive Robust and Adaptive

Control 1 Control 2 Control 3

AFTERNOON BREAK
MORNING BREAK

LUNCH BREAK

3B-05 3C-05 3D-05

M1
Pattern Recognition 1 Pattern Recognition 2 Pattern Recognition 3

3B-06 3C-06 3D-06

M2 Wireless and Wireless and Wireless and
Telecommunications 1 Telecommunications 2 Telecommunications 3

3B-07
3C-07 3D-07
M3 Bioinformatics and
Video Processing 1 Video Processing 2
Biomedical

3B-08 3C-08 3D-08

M4 Sensors and their Power Devices and Power Devices and
Applications Circuits 1 Circuits 2

3B-09 3C-09 3D-09

M8 Speech Recognition and Circuit Analysis and Circuit Analysis and
Extraction Design 1 Design 2

3C-10 3D-10
3B-10
M10 Power Systems Machine Modelling and
Watermarking
Harmonics Simulation

Tencon 2005 Page 17

Reviewers for Tencon 05
Name Affiliation
Adam Thompson MTData
Adel Ben Mnaouer NTU Singapore
Ajit Chaturvedi Indian Institute of Technology Kanpur, India
Alexander Kist RMIT University
Ali Agoub University of Gezira
Ali Al-Qayedi Etisalat University College
Ali Dawood Etisalat College of Engineering
Ali Memon Etisalat University College
Amin Sadik QUT
Ananda Kusuma University of Melbourne
Anant Malewar MoTECH Software
Andres Rojas Swinburne University of Technology
Andrzej Szymanski AGH University of Science and Technology
Andy Cockburn University of Canterbury
Angiras Varma University of Melbourne
Anna Lim University of Western Australia
Arafat Al-Dweik Etisalat University College
Arek Dadej University of South Australia
Aruna Jayasuriya University of South Australia
Asri Shaheem WATRI
Aun Haider University of Canterbury
Bartek Wydrowski California Institute of Technology
Behrooz Rohani Western Australian Telecommunications Research Institute
Beong Goo Ahn Hongik University
Bill Lloyd-Smith School of Computer Science and Information Technology, RMIT University
Bin Chen Nanyang Technological University
Binh Ngo University of South Australia
Biplab Sikdar Rensselaer Polytechnic Institute
Boon Hee Soong Nanyang Technological University
Boon Loong Ng University of Melbourne
Brent Martin University of Canterbury
Brooke Shrader University of Maryland
Bulent Sankur Bogazici University
Byungjin Chun University College Cork
Catherine Edwards Telstra
Chae-Woo Lee Ajou University
Chandra Athaudage University of Melbourne
Chandratilak Liyanage Massey University
Charles Begg University of Canterbury
Chee-Wah Tan Nanyang Technological University
Chen Khong Tham National University of Singapore
Choi Look Law Nanyang Technological University
ChorPing Low Nanyang Technological University
Christopher Leckie University of Melbourne
D. Manjunath Indian Institute of Technology,Bombay
Daeyoung Kim Information and Communications University
Dalma Novak Pharad
David Grayden Bionic Ear Institute
David Hayes Queensland University of Technology
Deepak Nayak IIT-Bombay
Donald Bailey Massey University

Tencon 2005 Page 18

Name Affiliation
Doo-seop Eom Korea University
Emre Aktas Hacettepe University
Enn Vinnal Telstra
Eric Lehmann University of Western Australia
Eric Wong City University of Hong Kong
Fabrizio Talucci i2telecom
Fawaz AL-Qahtani RMIT University
Felix Hartanto The Chinese University of Hong Kong
Feng Shu University of Melbourne
Firas Al-Ali Massey University
Franco Davoli CNIT
Gavin Baker University of Melbourne
Graham Goodwin The University of Newcastle
Greg Cresp WATRI
Gregory Day Curtin University of Technology
Grenville Armitage Swinburne University of Technology
Ha Duong University of South Australia
Hai Le Vu University of Melbourne
Hamza Ozer Turkish National Research Institute of Electronics and Cryptography
Heung-No Lee University of Pittsburgh
Hong Ren Wu RMIT University
Hong Seong Park Kangwon National University
Hongtao Tian Nanyang Technological University
Horace King Swinburne University of Technology
Huseyin Uzunalioglu Bell Labs, Lucent Technologies
Ian Holland WATRI
Irena Atov Swinburne University of Technology
Izhak Rubin University of California at Los Angeles
Jae-Hyun Kim Ajou University, South Korea
Jason But Swinburne University of Technology
Jayalakshmi Nambiar IIT-Bombay
Jeong-dong Ryoo ETRI
Jeonghoon Mo Information and Communications University
Jerzy Konorski Gdansk University of Technology
Ji Li University of Melbourne
Jill Slay University of South Australia
Jinwoo Jung Sangmyung University
Jitae Shin Sunkyunkwan University
Jong-Mu Choi Ajou University
Jun Guo University of Melbourne
Jusak Jusak RMIT University
Kahtan Mezher Etisalat University College
Kannan Govindan IIT-Bombay
Kannan Perumal Honeywell Technology Solutions Lab
Katrina Neville RMIT University
Kevin Lin RMIT University
Kihyung Kim Ajou University
Kiki Ren Telstra Research Laboratories
Kim Gribbon Massey University
Kim Jai-Hoon Ajou University
Kin Choong Yow Nanyang Technological University, Singapore
King-Tim Ko City University of Hong Kong
Krzysztof Pawlikowski University of Canterbury

Tencon 2005 Page 19

Name Affiliation
Kyeong-Deok Moon Electronics and Telecommunications Research Institute
Kyungran Kang Ajou University
Lachlan Andrew University of Melbourne
Lau Yuu Seng RMIT University
Lei Shang RMIT University
Lihai Liu City University of Hong Kong
Linda Davis Macquarie University
Mahmoud Al-Qutayri Etisalat University College
Majid Al-Dabbagh RMIT University
Malin Premaratne Monash University
Mark Halpern National ICT Australia (NICTA)
Markus Markgraf DLR
Matthew Peacock The University of Sydney
Mayank Keshariya University of Canterbury
Milosh Ivanovich Telstra Research Laboratories
Minh Tran Swinburne University of Technology
Mitch Thornton Southern Methodist University
Mohammed Al-Mualla Etisalat University College
Myung Hoon Sunwoo Ajou University, Korea
Nazar Ali Etisalat University College
Neville Churcher University of Canterbury
Neville Watson University of Canterbury
Oliver Yang University of Ottawa
Parampalli Udaya University of Melbourne
Patrick Hu University of Auckland
Paul Fitzpatrick Telstra Research Laboratories
Peter O'Shea Queensland Universtiy of Technology
Pramudi Suraweera University of Canterbury
Preeti Rao IIT-Bombay
Qassim Nasir University Of Sharjah
Qiang Fu Swinburne University of Technology
Rafal Stankiewicz AGH University of Science and Technology
Ramakrishnan Mukundan University of Canterbury
Ramesh Rayudu Massey University
Ray Hunt University of Canterbury
Rebecca Wang RMIT University
Richard Harris Massey University
Robert Suryasaputra RMIT University
Roger Browne Massey University
Ruili Wang Massey University
S. N. Merchant IIT Bombay
Saleh Al-Araji AC
Sam Yu Ajou University
Sammy Chan City University of Hong Kong
Sanghoon Lee Yonsei University
Sanjay Bose Nanyang Technological University
Sebastian Zander Swinburne University of Technology
Seedahmed Mahmoud RMIT University
Seong-Lyun Kim Yonsei University
Sergio Palazzo University of Catania
Seung-Woo Seo Seoul National University, Korea.
Shihab Jimaa Etisalat College of Engineering
Shyam Parekh Lucent Technology
Siow Yong Low Curtin University of Technology

Tencon 2005 Page 20

Name Affiliation
Son Young Sung ETRI
Steven Gordon University of South Australia
Subhas Mukhopadhyay Massey University
Sudhanshy Shekhar Jamuar Universiti Putra Malaysia
Sunghyun Cho Samsung Advanced Institute of Technology
Sungwoo Tak Pusan National University
Suresh Venkatachalaiah RMIT University
Suyong Eum RMIT University
Taka Sakurai University of Melbourne
Tasso Athanasiadis RMIT University
Tee Hiang Cheng Nanyang Technological University
Thaya Thanabalasingham The University of Melbourne
Thuy Nguyen Swinburne University of Technology
Tony Cui University of Melbourne
Tracy Camp Colorado School of Mines
Vasil Hnatyshin Rowan University
Walid Mahmoud University of Baghdad
Warren Harrop Swinburne University of Technology
Warwick Irwin University of Canterbury
Wayne Griffiths WATRI
Weihua Wang University of Melbourne
Wende Zhong Nanyang Technological University
Wendong Xiao Institute for Infocomm Research
William Zhiyong Liu University of Canterbury
Xia Li University of South Australia
Xiang Gui Massey University
Yatindra Nath Singh Indian Institute of Technology Kanpur
Yinghua Guo University of South Australia
Yong Liaw University of South Australia
Yongguang Zhang HRL Laboratories, LLC
Yoohwan Kim University of Nevada, Las Vegas
Young-Bae Ko Ajou University
Younglok Kim Sogang University
Youngsu Chae Yeungnam University, Korea
Youngyong Kim Yonsei University
Yun Li ChongQing University of Posts and Telecommunications of China
Zahir Hussain RMIT University

Tencon 2005 Page 21

Tencon 2005 Page 22
Guest Speakers
Tencon'05 is proud to announce the following keynote and invited speakers at our Conference:
Keynote Speaker: Dr Radia Perlman (Sun Microsystems)
Radia Perlman's work has had a profound effect on the world of
networking. She designed the spanning tree algorithm used by
bridges, and many of the key algorithms that make link state
protocols (IS-IS, OSPF) robust, manageable, and scalable. Her
work has also been seminal in routing security and tangible
computing. Recent contributions include secure digital shredding
of data, strong password protocols, transparent routing, scalable
PKI-based cross-organizational authentication and authorization,
and analysis and redesign of IKE (the authentication handshake
for IPsec). One of her missions is to get people to think critically
about networking, rather than just memorizing the details of what
happens to be currently deployed, or believing everything they
hear or see in print. Titles of recent talks she has given include "How to build an insecure system
out of perfectly good cryptography", "Things we all know about network protocols that aren't true",
and "Miss Manners Meets the IETF".
She is currently Distinguished Engineer at Sun Microsystems Laboratories. She is also a series
advisor for Prentice Hall, and serves on both the routing and security directorates of IETF.
She is the author of "Interconnections: Bridges, Routers, Switches, and Internetworking Protocols",
and coauthor of "Network Security: Private Communication in a Public World". Both books are
popular both with engineers and as university textbooks. She has taught graduate and
undergraduate level courses at Harvard, MIT, and University of Washington. Holding
approximately 70 patents, she was named Silicon Valley Intellectual Property Law Association's
2004 Inventor of the year.
Radia Perlman has a PhD in computer science from MIT and an honorary doctorate from KTH, the
Royal Institute of Technology, Sweden.

Mythology and Folklore of Network Protocols

It's natural to assume that network protocol design is a well-known science, where the designers of
today's standards take care to understand the tricks and pitfalls learned from previous protocols.
This talk dispels this and other myths. It is intended to be provocative, making people question the
things people assume are true; instructive, giving hints as to how to avoid some of the problems in
future protocols; and inspirational, convincing students that there are ample opportunities to make
contributions.

This talk discusses wrong turns that have been made, such as what necessitated the invention of
bridges, and what caused IP multicast to be unimplementable. It also talks about how a protocol,
even one “proven correct”, can go horribly wrong, such as the unstable ARPANET protocol for
distributing routing information. It talks about "obvious" tricks such as version numbers that even
today protocol designers insist on misusing. And it covers some of the areas in which research is
most needed.

Tencon 2005 Page 23

Keynote Speaker: Professor Rod Tucker (University of Melbourne)
Professor of Electrical Engineering, Head (1990-1992, 1996), Department
of Electrical and Electronic Engineering, and Director, Photonics Research
Laboratory, University of Melbourne; Director, ARC Centre for Ultra-
Broadband Information networks; Director - Melbourne, Australian
Photonics Cooperative Research Centre (1992-1998).

Fellow, Australian Academy of Science.

Fellow, Australian Academy of Technological Sciences and Engineering.
Fellow, Institute of Electrical and Electronics Engineers (IEEE).
Fellow, Institution of Engineers, Australia (IEAust).

Awards and Prizes

Harkness Fellowship, Commonwealth Fund of New York, 1975.
IEEE Lasers and Electrooptics Society, Distinguished Lecturer award, 1995.
Fisk Prize, Institution of Radio and Electronics Engineers, Australia, 1970.
M.A. Sargent Medal, awarded in 1995 by IEAust., for outstanding contributions to Electrical
Engineering.
Australia Prize, awarded by the Australian Government in 1997, for outstanding contributions to
Telecommunications.

Towards an Optical Internet

The Optical Internet is a vision that achieved prominence during the boom times of the late 1990’s.
This vision of a near-infinite-bandwidth network using advanced photonics was underpinned by a
surge in optimism about the capabilities of photonic technologies. With the passing of the boom, a
new realism has emerged about the role of photonics in optical networking. The vision of an
Optical Internet is not dead, but there now a clear understanding in the telecommunications
industry that the Optical Internet will become a reality only when it can provide a cost-effective
alternative to present-day technologies.

In this talk, I will give a personal perspective on some of the photonic technologies, circuits, sub-
systems, and architectures that will be needed to implement a cost-effective Optical Internet. I will
describe a number of evolutionary scenarios that may lead to increased use of photonic switching
and signal processing in the network, and compare the cost-effectiveness of a number of optical
network architectures. I will show how the relative costs of different architectures depend on the
cost of key photonic technologies and the total network capacity.

Tencon 2005 Page 24

Invited Speaker: Professor fred harris (San Diego State University)
fredric j. harris holds the CUBIC Signal Processing Chair of the
Communication Systems and Signal Processing Institute at San Diego
State University where since 1967 he has taught courses in areas
related to Digital Signal Processing and Communication Systems. He
has extensive practical experience in communication systems, high
performance modems, sonar and advanced radar systems and high
performance laboratory instrumentation. He holds a number of patents
on digital receiver and DSP technology and lectures throughout the
world on DSP applications. He consults for organizations requiring high
performance, cost effective DSP solutions.
He is well published and has contributed to a number of books on DSP.
In 1990 and 1991 he was the Technical and then the General Chair of
the Asilomar Conference on Signals, Systems, and Computers that
meets annually in Pacific Grove, California. He is the 2003 Technical Chair of the Software Defined
Radio Conference. In 2003 he became a Fellow of the IEEE and was cited for contributions of DSP
to communications systems. His education includes a Bachelor's Degree in EE from the
Polytechnic Institute of Brooklyn (1961), a Master's Degree in EE from San Diego State University
(1967) and Ph.D. work at the University of California, San Diego (1968-1973).
He is the traditional absent-minded professor and drives secretaries and editors to distraction by
requesting lower case letters when spelling his name. He roams the world collecting old toys and
slide-rules and riding old railways.

Signal Processing In Telecommunications

Many years ago a mentor questioned why I was pursuing Digital Signal Processing. He
commented, “All you can do with DSP is process audio, and why would you want to do that? I’m
glad I ignored his poor advice. I also remember sitting in a graduate course listening to an
instructor describe some aspect of optimum detection and thinking to myself, “It will never happen:
this is too complicated a process to ever see the light of day.” Ok, so I’m not right all the time! We
certainly have come a long way haven’t we? I am, as many of us are, immersed in DSP as an
educator, a practitioner, a researcher, a benefactor, or at least as a consumer. It is fun to pose to
the following questions: Where did it all start? Where is it now? Where is it going?

One thread in the sequence “where did it start, where is it now” is seen in the following. It was in
1965 that James Cooley and John Tukey published their paper “An algorithm for the machine
calculation of complex Fourier series" Who would ever have predicted that their Fast Fourier
Transform (FFT), an algorithm designed to compute spectra, would become the core of a
ubiquitous modulation and demodulation scheme known as Orthogonal Frequency Division
Multiplexing (OFDM). It is everywhere! It is used in Local Area Networking (LAN, IEEE 802.11a,
802.11g), in Digital Audio Broadcasting (DAB), in Digital Video Broadcasting (DVB), in Wide Area
Networking (WAN, IEEE 802.11n, 802.16a), in Asynchronous Digital Subscriber Lines (ADSL), and
in Multiple Input Multiple Output (MIMO).

This talk will be a nostalgia trip through the early days of DSP mixed with a report of the current
state of DSP in modern communications topped by a bit of crystal ball prediction of where DSP is
going.

Tencon 2005 Page 25

Invited Speaker: Dr Malin Premaratne (Monash University)
Malin Premaratne received the BSc(Maths.) and BE(Elec.) with first
class honours from the University of Melbourne, Australia, in 1995
and Ph.D from the University of Melbourne, Australia, in 1998. From
1998 to 2000, he was with the Photonics Research Laboratory, a
division of the Australian Photonics Cooperative Research Centre
(APCRC), the University of Melbourne, Melbourne, Australia, where
he was the co-project leader of APCRC Optical Amplifier Project.
During this period he worked with Telstra, Australia and Hewlett
Packard, USA through the University of Melbourne. From 2000 - 2003
he was involved with several leading startups in photonic area either
as an employee or a consultant.
During this period, he has also served in the editorial boards of
SPIE/Kluwer publishers and Wiley publishers in optical communications area. From 2001-2003, he
worked as the Product Manger (Research and Development) of VPIsystems Optical Systems
group.
Since 2000, he has been appointed as a Senior Associate/Senior Fellow of the University of
Melbourne, Australia. Since 2001 he also serve as the Chairman of IEEE Lasers and Electro-
Optics Society in Victoria, Australia.
Dr Premaratne is a senior member of IEEE and since 2003 leads the research program in high
performance computing applications to complex system simulation at Advanced Computing and
Simulation Laboratory (AXL) at Monash University, Australia. He has published over 75 research
papers in the areas of semiconductor lasers, EDFA and Raman amplifiers, optical network design
algorithms and numerical simulation techniques. His current research interests include the areas of
optical network design, analysis and configuration methods, simulation of optical and
electromagnetic interaction with biological and other turbid substances, grid and cluster computing
and realistic scene rendering in computer graphics using optical physics based methods.

Photonic Transport in Biological Tissue - Application of Optical

Communications Techniques to Bio-Sensing
In this talk I shall discuss applications of photon transport in biological tissue for non-invasive
sensing and imaging work. Much emphasis will be given to using already available tools in the
telecommunications area for such work. I shall also highlight some of the key fundamental
problems and challenges that need to be overcome to design and build next generation
biophotonic devices.

Tencon 2005 Page 26

Invited Speaker: Mike Schwartz (Executive Director, New Generation Systems, Telcordia
Technologies)
Mr Schwartz is sponsored by ATNAC 2005
Mike Schwartz is Executive Director in Telcordia Technologies New
Generation Systems organization. He is responsible for identifying the
impact of technology, regulation, and competition on Telcordia's new
software products and solution offerings. His work influences product and
solutions investments, marketing plan development, competitive response,
distribution and alliance strategies, and international strategies.
Mr. Schwartz's career spans over 37 years in the telecommunications
industry. Mr. Schwartz has led organizations responsible for the
development of industry technical requirements and standards, and
organizations responsible for the development and execution of
conformance and interoperability test capabilities. He was responsible for
the implementation of a new business within Telcordia to bring Telcordia's
technical services to the telecommunications equipment supplier market
place.
Mr. Schwartz received his BS (EE) from Rensselaer Polytechnic Institute in New York State, and
his MSEE from the University of California-Berkeley.
Mr. Schwartz is a member of the Tau Beta Pi Engineering Honor Society, and a Senior Member of
IEEE. He holds a seat on the Council of the World Telecommunications Congress and
International Symposium on Services and Local accesS (ISSLS) and is a former Chairman of
the ISSLS Council and its financial board.

Triple Play: Converge on the Screen, not on the Bill

Triple Play offerings today are all about discounting packages of voice, video, and data services,
but that's not where the future lies. The spread of broadband access and IP service platforms
allows operators to converge services where the customer uses them, at the communications
device.

Moreover, communications devices increasingly come with visual displays, or screens. With the
addition of video services to make the Triple Play, even the Home TV can become a central
communications device, where voice, data and video services are accessed and used. This kind
of convergence can transform the entire customer experience. This presentation will provide
examples of how services can be converged, and how such convergence can be used to
differentiate operator offerings on more than discounted price.

Tencon 2005 Page 27

Invited Speaker: Professor Ray Jarvis (Monash University)
Ray Jarvis completed his BE (Elec.) and Ph.D. (Elec.) at the
University of Western Australia in 1962 and 1968, respectively.
After two years at Purdue University he was appointed to the
Australian National University (ANU) to teach Computer Science.
He was instrumental in establishing the Department of Computer
Science at ANU and was its first Head of Department. He joined
Monash University in 1985 and established the Intelligent Robotic
Research Centre in 1987. He has been the Centre's Director
since then. His research interests are in Computer Vision and
Robotics. Both robotic manipulators and mobile robots of various
sizes have been part of his interests. Most recently he has been working on autonomous and tele-
operated vehicles and vessels including a tracked vehicle, several boats, an amphibious craft, an
excavator and a half scale Russian built Martian Rover. He has also recently developed a
research interest in Humanoid Robotics. From an AI perceptive his main interests are in Computer
Vision and Path Planning. He is Director of the recently established Australian Research Council
Centre for Perceptive and Intelligent Machines in Complex Environments.

Intelligent Robots: The Small, the Big and the Many

Intelligent Robots come in a variety of shapes and sizes and can operate individually and in
swarms. This paper will cover the underlying navigation principals for such machines and provide
a number of case studies to illustrate how these principals can be integrated.

Tencon 2005 Page 28

 1B-01
Session Title: Traffic and Resource Management 1
Chairperson: Greg Egan
Day: Tuesday
Date: 22nd November 2005
Time: 11:20AM – 1:00PM
Room: M1

1B-01.1 1568961181 interpersonal comparisons of utility and the bargaining

problem are investigated. In the aspect of the social welfare
An Effective Hotspot Cell Management Scheme (total utility), the results also indicate that the Nash solution
Using Adaptive Handover Time in 4G Mobile of the bargaining problem outperforms the equilibrium point
Networks and the solution of the interpersonal comparisons of utility
due to its much higher total utility."
Dong Wook Kim
"We propose a hotspot cell management scheme using 1B-01.3 1568964014
adaptive handover time in 4G mobile networks. 4G
networks are expected to support various multimedia Evaluating the Restorability in WDM Networks
services over IP network and satisfy high spectral efficiency Based on Directional Cycle Decomposition
requirement with overcoming time and frequency selective Algorithm (DCDA)
fading due to the high speed of users. In cellular systems
including 4G networks, hotspot cells can be occurred when
Chuan-Ching Sue, JunYing Yeh, Shih-Chang Hsu,
available wireless resources at some location are not Chao-Yang Lee
enough to sustain the needs of users. The hotspot cell can "The Directional Cycle decomposition Algorithm (DCDA) is
potentially lead to blocked and dropped calls and cause one method to achieve full restoration and tolerate as many
degradation of the service quality. In the meanwhile, a band faults as possible in wavelength division multiplexing
of users enjoying multimedia services can move around in a (WDM) networks without the capability of wavelength
4G mobile network, which may generate heavy flows of conversion. This paper formulates the restorability
traffic load in the network. This situation can generate the computing in WDM network based on DCDA. Two methods
hotspot cell which has a short length of life at most a few are proposed to compute the restorability. One computes
minutes. In this paper, we propose a handover-based the restorability exactly using the information from the
scheme which can effectively manage hotspot cells in 4G original network and its corresponding G+. Another
mobile networks. The proposed scheme adopts hard computes approximately using the information only from the
handoff mechanism and dynamically controls the time point corresponding G+. The results indicate the restorability is
of handover based on the amount of load of the current decreasing very fast with more and more simultaneous
serving cell and the target cell. Adaptation of handover time faults. Moreover, the average error in the approximate
according to the amount of load effectively and flexibly can method is limited from 8.37% on ArpaNet to 2.94% on
manage the hotspot cell. And, through our hotspot cell Icosahedron."
management scheme, acceptable service quality can be
supported as users maintain graceful connections with the 1B-01.4 1568964791
network. In the simulation results, we find that our scheme
makes smaller number of hotspot cells and supports higher Analysis of Real-Time Multiple Source Video
service quality than compared schemes." Streaming via Wireless 5.8 GHz for Intelligent
Traffic Management System
1B-01.2 1568963458
Vishnu Monn Baskaran
A Game Theoretic Framework for Call Admission
"This paper evaluates the performance of live compressed
Control in CDMA Mobile Multimedia Systems with motion image transmission in coherence with the
Traffic Asymmetry between Uplink and Downlink development of the Multiple Video Streaming and
Jenjoab Virapanicharoen, Watit Benjapolakul Compression (Multiple-VSCS) software, as applied for the
Intelligent Traffic Management system. Motion Images are
"The call admission control (CAC) for code-division captured, compressed and streamed real time via utilization
multiple-access (CDMA) mobile multimedia systems is a of the 5.8 GHz OFDM Outdoor Wireless LAN network.
very important engineering issues since it belongs to the Experiments were performed and analyzed to ascertain the
category of resource management and the radio spectrum reliability and potency of this application with improved
is a very scarce resource. Since efficiency is an important bitstream rates as low as 50kbps and compression ratio
performance issue in the mobile networks with multimedia rates as high as 1800:1, under various compression
services, the concept of fairness among services should parameters settings as well as different video formats. This
also be considered. Game theory can provide an paper also includes performance analysis of video stream
appropriate framework for formulating such fair and efficient rates with comparison into varying compression formats
CAC problem. In this paper, a selection method for fair- and Quantization Factors retrospective to Quality of Service
efficient threshold parameters of the CAC scheme in CDMA (QoS). Projections were also performed as behavioral study
mobile multimedia systems with traffic asymmetry based on on the effects of the available bandwidth with increased
the framework of game theory is proposed. The proposed video capture sources and streaming processes. Keywords:
scheme provides fairness and efficiency in terms of uplink H.263 compression codec, Outdoor OFDM 5.8 GHz WLAN,
and downlink signal-to-interference ratio (SIR) for the 2 Traffic Management System"
classes of traffic, and also guarantees the priority of class 0
calls over class 1 calls. Class 0 calls and class 1 calls are
viewed as two players of a game. The numerical results
show that there is an equilibrium point of the
noncooperative game for any given offered load. For the
cooperative game, the arbitration schemes for the

Tencon 2005 Page 29

1B-01.5 1568964827
PIPPON: A Physical Infrastructure-aware P2P
Overlay Network
Doan Hoang, Hanh Le, Andy Simmonds
"The scalability and efficiency of P2P systems are
determined largely by two factors: the degree of matching
between the P2P overlay and its underlying physical
network, and the efficiency of the P2P application routing
scheme. Most current P2P systems address one or the
other but not both in an integrated manner. This paper
proposes PIPPON, a two-layer architecture to address
these performance issues. At the overlay network layer, the
paper proposes a simple but effective Geo-LPM/Geo-
Partitioning scheme for constructing overlay that mirrors the
physical Internet. At the application routing layer, the paper
introduces a dynamic Key Tree that is built on the optimized
overlay to rout P2P messages efficiently. Simulation results
indicate that PIPPON is scalable and efficient, and holds
promise as a platform for large scale P2P applications."

Tencon 2005 Page 30

 1B-02
Session Title: Ad Hoc Networks
Chairperson: Alexander Kist
Day: Tuesday
Date: 22nd November 2005
Time: 11:20AM – 1:00PM
Room: M1

1B-02.1 1568963601 1B-02.4 1568965270

CLA-QOS: A Cross-Layer QoS Provisioning An Intrusion Detection System for Suburban Ad-
Approach for Mobile Ad-hoc Networks hoc Networks
Hongtao Tian, Sanjay Bose, Choi Look Law, Muhammad Islam, Ronald Pose, Carlo Kopp
Wendong Xiao "Due to the nature of the wireless media, ad-hoc wireless
"The increasing popularity of the multimedia applications in networks are vulnerable to various attacks. There are
mobile ad hoc networks (MANETs) has led to research on security protocols that prevent unauthorized nodes from
QoS support for these applications. However, this is difficult accessing the network through authentication. Secrecy of
in such networks because of factors such as shared and information is provided through encryption. However these
variable wireless bandwidth, limited power supply and protocols cannot detect if any member of the network
possible mobility of the nodes. Although many QoS degrades the network performance due to misbehavior.
provisioning schemes have been presented, few have Therefore an intrusion detection system (IDS) is required
provided the end-to-end QoS assurance from the point of that monitors what is going on in the network, detects
view of cross layer optimization. In this paper, a cross-layer misbehavior or anomalies based on the monitored
approach to QoS provisioning (CLA-QOS) is proposed to information and notifies other nodes in the network to take
provide such QoS support. Simulation results show that the necessary steps such as to avoid or punish the
CLA-QOS approach provides QoS guarantees in terms of misbehaving nodes. In this paper we propose an IDS,
the in-time packet delivery ratio of packets. Simulation referred to as the SAHN-IDS, suitable for multi-hop ad-hoc
studies supporting this are provided." wireless networks like a SAHN (Suburban Ad-hoc Network).
SAHN-IDS detects misbehavior based on nodes getting an
1B-02.2 1568964415 unfair share of the transmission channel. It also detects
anomalies in packet forwarding, such as intermediate nodes
Modifying AODV for Efficient Power-Aware dropping or delaying packets. Unlike most IDSs for
Routing in MANETs detecting anomalies in packet forwarding, SAHN-IDS does
rely on overhearing packet transmissions of neighboring
Chee-Wah, Tan Sanjay Bose
nodes, since that is ineffective in networks where nodes use
"Unlike infrastructure based wireless networks, nodes in different transmission power, different frequency channels
mobile ad hoc networks (MANETs) will also have to act as and directional antennas for different neighbors. Moreover,
intermediate nodes for data routing. These additional unlike most IDSs, most of the thresholds in SAHN-IDS are
transmissions will further drain the power resources of set dynamically. We show the effectiveness of SAHN-IDS
mobile nodes which are anyway likely to be running on through simulations."
limited battery power. This paper proposes a Power-Aware
Ad hoc On-demand Distance Vector routing protocol (PAW- 1B-02.5 1568964898
AODV) for such power limited applications. This uses the
overall power resources of the system more efficiently than "Ranger, a Novel Intrusion Detection System
the standard AODV protocol by choosing routes based on a Architecture for Mobile Ad Hoc Networks"
power-based cost function. We implement our proposed Yinghua Guo, Steven Gordon
PAW-AODV protocol by suitably modifying AODV and study
it through simulations using QUALNET. Simulation results "The proliferation of wireless communication and mobile
show that PAW-AODV is able to provide more throughput computing is driving the emergence of Mobile Ad hoc
than AODV over the life of the system and can make better Networks (MANETs) with wide application ranges from
use of the limited battery power available." civilian environment to military communication. However,
securing MANETs is a highly challenging issue due to their
1B-02.3 1568965127 inherent characteristics. Intrusion detection is an important
security mechanism, but little effort has been directed
A Heuristic for low power QoS routing in ad hoc towards efficient and effective architectures for Intrusion
networks Detection System (IDS) in the context of MANETs. We
investigate existing IDS architecture design issues, and
Somnath Ghosh, Stacey Osbrough, Samar Singh
propose a novel mobile agent based IDS architecture that
"For an important class of ad hoc wireless networks, has each node implementing basic IDS functions, while
conserving power is a major concern. All nodes in an ad ranger agents roam the network executing more advanced
hoc network participate in routing and forwarding of IDS functions. This is suited to MANETs because it avoids
packets. Many ad hoc networks support applications which the single point of failure problem, minimises
require network connections with guaranteed quality of communication overheads at the same time as providing up
service (QoS). In this paper a simple routing scheme is to date information for intrusion decisions."
presented for controlling the network topology so that QoS
routes are provide while keeping transmitter power at the
nodes close to minimum."

Tencon 2005 Page 31

 1B-03
Session Title: Image Processing 1
Chairperson: Donald Bailey
Day: Tuesday
Date: 22nd November 2005
Time: 11:20AM – 1:00PM
Room: P1

1B-03.1 1568963922 1B-03.4 1568964757

A Rotational Motion Estimation Method for Digital Feature-Oriented Coupled Bidirectional Flow for
Image Stabilization Image Denoising and Edge Sharpening
Joon Lyou, Ho Dong Seok, Hwy Kuen Kwak Shujun Fu, Qiuqi Ruan, Yuliang Geng, Wenqia Wang
"A simple and robust digital image stabilization system "In this paper, a new type of diffusion process that
subjected to a rotational and translational motion is devised. simultaneously denoises and sharpens images is
It is shown that single rotation model can accommodate a considered. We presents a feature-oriented coupled
combined rotational and translational motion. Thanks to this bidirectional flow process, where, according to image
simplification, the present digital image stabilization(DIS) features such as edges, textures, and fine parts, it can
system works well via estimation of rotation center and switch from a forward diffusion to a backward (inverse) one
angle between image frames, instead of estimation of both along the normal directions to the isophote lines (edges),
the rotational and translational motion parameters as found while a forward diffusion is performed along the tangent
in the previous DIS schemes. Hence, the computational directions. To eliminate the conflict between the backward
burden of our algorithm is considerably less, because it and the forward force, we split them into a coupled scheme.
does not require the time consuming parameter searching, In order to enhance image features the nonlinear diffusion
while showing a comparable performance to the previous coefficients are locally adjusted according to the directional
ones." derivatives of the image. Experimental results demonstrate
that our algorithm can substantially enhance features on
1B-03.2 1568964731 denoising smoother areas of the image."
Medical Image Processing: From Lossless to 1B-03.5 1568964775
Lossy Compression
Semi-supervised Segmentation of Textured
Oh Tick Hui, Heng Siong Lim, Shen Yee Pang Images by Using Coupled MRF Model
"Compressed image will always reduce the image fidelity.
This is especially so if the image is compressed at lower bit
Yong Xia, Dagan Feng, Rongchun Zhao
rates, which most of the time cannot be tolerated. "Markov Random Field (MRF) is extensively used in model-
Compression is necessary in medical field due to the limited based segmentation of textured images. In this paper, we
transmission bandwidth and storage capacity. In this paper, propose a coupled MRF model and adopt the MAP-MRF
performance of the controversial lossy compression on framework to solve the semi-supervised segmentation
medical modalities using the new JPEG2000 and the more problem. The observed image and the desired labeling are
conventional JPEG is being studied. The common objective characterized by the Conditional Markov (CM) model and
measure, peak signal-to-noise ratio (PSNR) is definitely not the Multi-Level Logistic (MLL) model, respectively. The
adequate as a perceptual quality measurement. Therefore, parameters of CM models are estimated as texture
subjective measures such as picture quality scale (PQS) features, and contextual dependent constraints are posed
and mean opinion score (MOS) are used as the benchmark to the object function by the MLL model. Different from the
instead, along with the statistical and frequency measures existing methods, the two MRF models are mutually
for the image characteristics evaluation. Three types of dependent in our approach and therefore texture features
medical modalities are used X-ray, Ultrasound and and the labeling must be optimized simultaneously. To this
Computed Tomography (CT). From these various end, a step-wised optimization scheme is presented to
measurements, the general acceptable compression bit rate achieve a suboptimal solution. The proposed algorithm is
for each of the medical modality is determined." compared with a simple MRF model based method in
segmentation of Brodatz texture mosaics. The experimental
1B-03.3 1568964744 results demonstrate that the novel approach can
differentiate textured images more accurately."
CFA Demosaicking with Improved Colour Edge
Preservation
Sharmil Randhawa, J.S. Jimmy Li
"Colour edges are sharp boundaries between two distinct
colours. Most exisiting demosaicking methods depend on
assumptions which will be invalid at edges. In this paper,
we apply a method that avoids interpolation across an edge
by extrapolating estimates in four different directions. In
order to improve colour edge preservation, we propose a
novel classifier to select the best estimate in conjunction
with its surrounding pixels. It has been shown that our
method outperforms visually and quantitatively with image
quality measures, when compared with other existing
methods."

Tencon 2005 Page 32

 1B-04
Session Title: IT Applications 1
Chairperson: Horace King
Day: Tuesday
Date: 22nd November 2005
Time: 11:20AM – 1:00PM
Room: P3

1B-04.1 1568964598 optimal resource allocation problems in a module software

during testing phase: 1) minimization of the remaining faults
Automated Translation of Behavioral Models when a fixed amount of testing-effort is given, and 2)
using OCL and XML minimization of the required amount of testing-effort when a
Sabnam Sengupta, Ananya Kanjilal, Swapan specific reliability requirement is given. Several useful
optimization algorithms based on Lagrange multiplier
Bhattacharya
method are proposed. Furthermore, we present the
"The UML diagrams present different but partly overlapping sensitivity analysis on these allocation problems in order to
views of the same system and hence design risks due to determine which of the parameters affects the system most.
inconsistencies among diagrams is inevitable. Automated Finally, a numerical example is evaluated to validate and
translation of behavioral diagrams is one way of ensuring show the effectiveness of the proposed approach."
consistency among the different models that can take us
one step forward towards formalizing behavioral 1B-04.4 1568964866
specifications of the system. In this paper based on the
dependencies between some of the commonly used A BDL-Based E-Commerce Transaction
diagrams, we have proposed a methodology for translating Architecture
automatically a set of behavioral diagrams (Sequence to Lakshmi Narasimhan, Renyi Zhou
State Transition Diagrams). As this translation is
automated, the consistency among the diagrams is "This paper describes extensions to a Behavioral
automatically checked. Our approach combines both OCL Description Language (BDL), which was originally proposed
and XML technologies. Based on the OCL constraints, we to characterize concurrent behavior of simple objects and a
here propose a methodology for automated derivation of group of objects. One of the novelties of this paper is its
statechart diagram from a set of sequence diagrams with application to the field of E-Commerce transaction systems.
the help of XML." Based on the BDL, we propose new concepts, namely,
transaction patterns and transaction architectures, which
1B-04.2 1568964740 have event-based semantics to describe large-scale
transaction systems. Furthermore, the transaction
A Dual-Spiral Reengineering Model for Legacy architecture is introduced as a unified medium for specifying
System and verifying distributed, heterogeneous and complex E-
Commerce transaction processes. It is also illustrated as a
Xiaohu Yang, Lu Chen, Xinyu Wang, Jerry Cristoforo
powerful modeling technique which is easy-to-use, flexible
"Computer Technology has been developing for a long and promotes re-usability."
time; more and more systems become outdated. As a
result, Legacy System Reengineering becomes a critical 1B-04.5 1568964970
effort in many business fields. There are some existing
reengineering models, such as Byrne Model and Masking approach to secure systems from
Evolutional Model. They have resolved some problems in Operating system Fingerprinting
the reengineering process, but are not applicable to some Surbhie Kalia, Maninder Singh
scenario. This paper presents a dual-spiral reengineering
model, which performs as a cyclic approach, and it could The system vulnerabilities are most precious thing for
avoid the issues that those models have, making the attackers. One such vulnerability is OS of the system which
reengineering process more applicable. This model has can be easily fingerprinted. This paper basically deals with
been successfully applied to a legacy financial software discussion about the primary phases that every OS
system reengineering. We will discuss the issues in the fingerprinting tool uses to detect the OS of Remote System.
Byrne and Evolutional models first, and then present the The techniques used by popular tools Nmap and Xprobe2
detail description of Dual-Spiral Reengineering Model which are presented here, in context to understand the security
is based on spiral model of software development and framework that is to be setup to play defensive against
enhancement. At last, we will introduce a reference case for them. Further some implementation examples are
our new model." illustrated for different OS to countermeasure against
Fingerprinting."
1B-04.3 1568964857
An Algorithm to Allocate the Testing-Effort
Expenditures Based on Sensitive Analysis
Method for Software Module Systems
Jung-Hua Lo
"To develop a good reliable software system, a project
manager must determine in advance how to effectively
allocate these resources. Due to the dynamic properties of
the system and the insufficiency of the failure data, the
accurate parameter values are hard to determine.
Therefore, the sensitivity analysis is often used in this stage
to deal with this problem. In this paper, we study two

Tencon 2005 Page 33

 1B-05
Session Title: Modulation & Coding 1
Chairperson: Graham Woods
Day: Tuesday
Date: 22nd November 2005
Time: 11:20AM – 1:00PM
Room: M2

1B-05.1 1568963925 1B-05.4 1568964745

Versatile Multiplexing Based on Hunting-Free A Robust Design of Reconfigurable
Framing Coder/Decoder Against Un-authorized Users
Yoshitaka Takasaki Chuan-Ching Sue, Jen-Fa Huang, Yao-Tang Chang,
"Application of multipled block codes (MBCs) for realizing Deng-Sheng Wang, Chao-Yang Lee
new, flexible and versatile multiplexing is investigated. First, "In this paper, an enhanced security mechanism is
the principles of MBC are overviewed to show its capacities presented against eavesdropping on spectral amplitude
of filterless clock recovery. Then it is shown that coding (SAC) optical code-division multiple-access
modification of simple frame structure of MBC line code can (OCDMA) access networks. The reconfigurable
be used for attaining hunting-free reframing for multiplexing coder/decoder (codec) which was constructed with Arrayed-
systems. Finally, the applicatility of hunting-free reframing to Waveguide-Grating (AWG), optical switch and additional
versatile multiplexing is investigated. An example of controller unit is used as the basic system to realize the
broadband ubiquitous subscriber distribution in terms of implementation of dynamic code matrix assignment. Instead
versatile multiplexing is discussed." of employing a huge code space, we adopt the spectral
amplitude coding (SAC) with unipolar M-sequence code
1B-05.2 1568964586 matrix for a specific signature address (coding) and its
match address codeword (decoding). By using the inherent
Exact Closed-Form Performance Analysis of cyclic properties of both AWG routers and M-sequence
MFSK Demodulator Based on FFT codes, the reconfigurable coding/decoding scheme is
Sungdon Moon developed to avoid the attack of un-authorized users even
with fewer code spaces. Furthermore, a weighted load
"In this paper, we investigate the BER performance of M- balance problem based on the power distribution on each
ary FSK modulation system using FFT-based demodulator wavelength is investigated, and then a dynamic
scheme. An explicit BER expression of the FFT reconfiguration policy is proposed to choose the best code
demodulator scheme is analytically derived in terms of a matrix assignment with the minimum weighted degree of
closed form and compared with a typical demodulation load balance. The evaluation results show that the dynamic
scheme, the non-coherent detection method. In addition, in reconfiguration policy can outperform a class of static
order to verify the derived results, the simulation results are policies in terms of two performance metrics, weighted
together compared. It is noteworthy that the performance of degree of load balancing and the number of register
M-ary FSK system with the FFT method is equal to that with shifting."
non-coherent detection."
1B-05.5 1568964891
1B-05.3 1568964676
Geometrical and Phasor Representation of Multi-
Improving Group Delay Measurement Accuracy level Amplitude-Phase Modulation Formats and
using the FM Envelope Delay Technique Photonic Transmitter Structures
Graham Woods, Douglas Maskell Dung Tran, Le Binh, Thanh Huynh, Ho Siang Tiong
"The FM envelope delay technique of measuring group
"We describe a geometrical and phasor representation of
delay is an attractive alternative to standard techniques
multi-level amplitude and phase modulation. Based on this
using equipment like the network analyzer. The theoretical
representation, the modulation of the lightwave carrier is
basis for this technique is more complicated though leading
described using single dual-drive Mach Zehnder
to some confusion about how to optimize the measurement
interferometric modulator. The generation of optical
accuracy. This paper discusses methods of improving
multilevel amplitude and phase constellations (oMASK or
accuracy when making group delay measurements using
oMPSK) is designed for effective photonic transmission. A
the FM envelope delay technique. The paper concentrates
number of photonic transmitter structures for the modulation
on two effects which are know to limit accuracy in this
schemes are briefly compared. The driving voltage levels
approach; the finite bandwidth of the test signal and device
are calculated and the phasor diagrams are developed to
mismatch. The potential severity of each source of
demonstrate the effectiveness of modulation schemes.
uncertainty is modeled from theoretically derived
Simulink models are developed to generate the advanced
expressions and computer simulations. Practical
modulation formats for optical fibre communications.
measurements made using a FM envelope delay, group
Experimental demonstrations will be illustrated."
delay measurement system are also provided to verify the
predicted trends. Adjusting the FM signal modulation index
is shown to be an effective method of minimizing error
depending on the characteristics of the device being
measured."

Tencon 2005 Page 34

 1B-06
Session Title: Time-Frequency Analysis and Wavelets 1
Chairperson: Le Binh
Day: Tuesday
Date: 22nd November 2005
Time: 11:20AM – 1:00PM
Room: M3

1B-06.1 1568963787 1B-06.4 1568964729

Application of Wavelet Array Denoising to Comparison of few wavelets to filter ocular
ESPRIT Algorithm artifacts in EEG using Lifting Wavelet Transform
Yanbo Xue Salwani Mohd. Daud
"ESPRIT algorithm has been widely used in direction-of- "We proposed a technique to remove ocular artifact (OA)
arrival (DOA) estimation problem. It degrades greatly in the from electroencephalogram (EEG) using Lifting Wavelet
presence of spatially correlated noises and closely spaced Transform (LWT). The objectives of OA correction are
DOA's, and/or at low SNR. Filters are often used to harness threefold 1) to reduce the OA magnitude, 2) to retain the
the performance degradation because they can recover the neural EEG potentials and 3) to make corrections at the
signal from its noisy observation. Though Wiener filter is the right time localization. Three basic steps involved are to
optimal filter, it requires the knowledge of second-order transform the EEG, hard thresholding the wavelet
statistics of the signals and noises, which is difficult to coefficients and the corrected EEG is obtained by inverse
obtain in non-wide-sense stationary signals and noises. transform these threshold coefficients. It is of paramount
Wavelet array denoising has been introduced to ESPRIT important to select a suitable wavelet and threshold to
method in this paper to provide a possible filtering method accomplish this task. In this investigation, few wavelets
in spatially correlated noises. The proposed approach were used and their effectiveness performances were
denoises the snapshots of each sensor in parallel and compared visually. A level-dependent threshold used was
applies conventional ESPRIT algorithm to the denoised suitable for both blinking and eyeball movements correction.
data matrix. Simulation results show that by using the From this study, we concluded that Haar and rbio3.3
proposed approach, we can gain advantages of estimation outperformed db4 and cdf4.4 wavelets, and had achieved
root mean square error (RMSE) reduction and resolution the objectives of OA removal."
enhancement."

1B-06.3 1568964702
Design and Implementation of Wavelet Filter
Using Switched-current Circuits
Qinchun Hu, Yigang He, Dixin Guo, Hongmin Li,
Meirong Liu
"The design and implementation of wavelet filters using
switched-current circuits are presented in this paper.
Wavelet filters play a key role in analog hardware
implementation of wavelet transform, which have been an
attractive option to achieve real-time performance because
that every signal is not suitable for digital processing. The
switched-current (SI) technique is a relatively new analog
sampled-data signal processing technique that aims to
replace switched-capacitors. Design techniques for
switched-current filters have been described. The key
feature of SI circuits from the viewpoint of wavelet transform
is that changing the dilation constant can be very precisely
controlled by clock frequency. Thus, a seed wavelet filter
can be implemented directly as SI filter circuits, with other
filers at different scales being simple dilations of it. The
desired transfer functions of wavelet filters are
approximated by Pade approximation which is able to
decompose the transfer function into rational form so as to
be conveniently implemented by SI circuits. Therefore,
arbitrary wavelet filters can be implemented by the method
proposed in this paper. Simulated results have
demonstrated the effectiveness of the proposed new
method for the implementation of wavelet filter."

Tencon 2005 Page 35

 1B-07
Session Title: Electronic Devices 1
Chairperson: Amin Sadik
Day: Tuesday
Date: 22nd November 2005
Time: 11:20AM – 1:00PM
Room: M4

1B-07.1 1568963793 tracked. The tracking algorithm integrated with a solar-

powered battery charging system has been successfully
Spiral Inductor Coupling on 0.25 um Epitaxial implemented on a low-cost PIC16F876 RISC-
CMOS Process microcontroller without external sensor unit requirement.
John Richard Hizon, Marc Rosales, Louis Alarcon, The experimental results with a commercial solar array
show that the proposed algorithm outperforms the
Delfin Jay IX Sabido
conventional controller in terms of tracking speed and
"In this study, the degree of coupling between two adjacent mitigation of fluctuation output power in steady state
inductors is examined. The inductors were implemented on operation. The overall system efficiency is well above 90%."
a 0.25 m epitaxial CMOS process. Coupling
mechanisms considered in this study were magnetic 1B-07.4 1568964855
coupling and substrate coupling. The amount of coupling is
determined by measuring the s21 parameter between the A Complete Isothermal Model for the Lateral
inductors implemented. The effectiveness of several Insulated Gate Bipolar Transistor on SOI
isolation structures reported in literature in reducing technology
coupling between inductors was evaluated. These isolation
structures include patterned ground shields and halo
Vasantha Pathirana
substrate contacts. From measured results, it is found that "While there are several analytical models dedicated to the
reducing magnetic coupling by implementing a diagonal vertical IGBT there is virtually no reliable model for the
configuration improves isolation by 5 dB. Furthermore, Lateral IGBT (LIGBT). LIGBTs are increasingly popular in
patterned ground shields are effective in improving the smart power and power integrated circuits, especially in
isolation between adjacent inductors." those applications where high voltage and high current
capabilities are required. Silicon on insulator (SOI) could be
1B-07.2 1568963903 the future technology for power devices and integrated
circuits due to its much improved device isolation. In this
Development of a conductive photoresist with a work we report for the first time a complete analytical model
mixture of SU-8 and HCl doped polyaniline for the SOI LIGBT based on semiconductor physics with
Uma Annaiyan, Kourosh Kalantar-Zadeh, John very few fitting parameters. The model is implemented in
the widely available circuit simulator PSpice. The model is
Quang Fang, Irena Cosic
assessed against finite element device simulations and
"A novel electrically conductive photoresist has been experimental results."
formulated to fabricate microcomponents. The developed
conductive photoresist is based upon SU-8 photopolymer, 1B-07.5 1568964880
an insulating negative-tone epoxy, in which protonically
doped polyaniline (PANI) nanoparticles have been Threshold Features of L-Band Linear Cavity
dispersed to enhance the electrical properties. The Multiwavelength Brillouin-Erbium Fiber Laser
characteristics of this new conductive photoresist have Mohammed Hayder Al-Mansoori Syed, Javaid
been studied via electrical measurements. The process for
preparing the conductive films from the combination of EB
Iqbal, Mohamad Khazani Abdullah, Mohd Adzir
(Emeraldine base) PANI, SU-8 and NMP (N-methyl-2- Mahdi
pyrrolidinone) will be presented. Different weight Investigation of L-band linear cavity multiwavelength
percentages (wt %) of SU-8 in the above combination have Brillouin-Erbium fiber laser source has been experimentally
been prepared and spin coated to form thin-films which studied. The effects of Brillouin pump power and
have been protonated with HCl. The conductivities of the wavelength on the threshold pump power are discussed in
thin-films were measured using a four point probe. The the paper. We found that there is a limited range for
highest conductivity achieved was approximately 1.6 S/cm Brillouin pump wavelength to achieve a lasing condition in
for the mixture of 10wt% of SU-8 25 with EB-PANI mixed the laser cavity. Low threshold power of 23.4 mW is
with NMP. The morphology of the thin-films was studied obtained from the linear cavity due to highly efficient from
using a Scanning Electron Microscopy (SEM)." both double-pass amplification in the Erbium gain medium
and also bidirectional generation of Brillouin Stokes in the
1B-07.3 1568964665 single mode fiber.
A Solar-powered Battery Charger with Neural
Network Maximum Power Point Tracking
Implemented on a Low-Cost PIC-microcontroller
Panom Petchjatuporn, Ms.Wannaya Ngamkhak,
Noppadol Khaehintung, Phaophak Sirisuk, Wiwat
Kiranon, Anatawat Kunakorn
"This paper presents the development of a maximum power
point tracking algorithm using an artificial neural network for
a solar power system. By applying a three layers neural
network and some simple activation functions, the
maximum power point of a solar array can be efficiently

Tencon 2005 Page 36

 1B-08
Session Title: Biomedical Engineering 1
Chairperson: James Cheong
Day: Tuesday
Date: 22nd November 2005
Time: 11:20AM – 1:00PM
Room: M6

1B-08.1 1568964867 1B-08.3 1568964985

Mapping of Genetic Algorithms for Protein Experimental Comparison of Initial Pool
Foilding onto Computational Grids Generation for Direct-Proportional Length-Based
Weiguo Liu, Bertil Schmidt DNA Computing
"Computational biology research is now faced with the Zuwairie Ibrahim, Mohd Saufee Muhammad, Yusei
burgeoning number of genome data. The rigorous Tsuboi, Osamu Ono, Marzuki Khalid
postprocessing of this data requires an increased role for "Previously, direct-proportional length-based DNA
high performance computing. In this paper, we present a computing (DPLB-DNAC) for solving weighted graph
framework to map hierarchical genetic algorithms for protein problems such as the shortest path problem, has been
folding problems onto computational grids. It has been proposed. After the initial pool generation, the initial pool is
designed to take advantage of the communication subjected to amplification, by polymerase chain reaction
characteristics of a computational grid. By using this (PCR), and finally, the output of the computation can be
framework, the two level communication parts of visualized by gel electrophoresis. In this paper, we give
hierarchical genetic algorithms are separated. Thus both more attention to the initial pool generation of DPLB-DNAC.
parts of the algorithm can evolve independently. This For this purpose, two kinds of initial pool generation
permits users to experiment with alternative communication methods, which are hybridization/ligation and parallel
models on different levels conveniently. Our experiments overlap assembly (POA) are evaluated. We found that both
show that it can lead to significant runtime savings on PC of the methods can be successfully employed to generate
clusters and computational grids." the initial pool generation of DPLB-DNAC, as supported by
results of actual experiments. However, POA is more
1B-08.2 1568964879 preferable than that of hybridization-ligation in term of
Gene Selection and Classification of Human population size and generation time."
Lymphoma from Microarray Data
1B-08.4 1568964991
Joarder Kamruzzaman
A Semi-automatic System for Measuring Tibial
"Experiments in DNA microarray provide information of Cartilage Volume
thousands of genes, and bioinformatics researchers have
analyzed them with various machine learning techniques to James Cheong, David Suter, Flavia Cicuttini
diagnose diseases. Recently Support Vector Machines "Osteoarthritis is a chronic and crippling disease affecting
(SVM) have been demonstrated as an effective tool in an increasing number of people each year. With no known
analyzing microarray data. Previous work involving SVM cure, it is expected to reach epidemic proportions in the
used every gene in the microarray to classify normal and coming years. Currently, there is strong interest in
malignant lymphoid tissue. This paper shows that, using developing a fully automated cartilage area/volume
gene selection techniques that selected only 10\% of the measurement method in the medical field to assist both
genes in `Lymphochip'' (a DNA microarray developed at pharmaceutical companies and medical professions in
Stanford University School of Medicine), a classification researching the disease. This paper describes the
accuracy of about 98\% is achieved which is a comparable development of a semi-automatic system for segmenting
performance to using every gene. This paper thus and measuring human knee cartilage volume from magnetic
demonstrates the usefulness of feature selection resonance imaging (MRI) scans. The cartilage volume
techniques in conjunction with SVM to improve its obtained from the semi-automated method has been
performance in analyzing Lymphochip microarray data. The benchmarked against the current gold standard (cartilage
improved performance was evident in terms of better volume from manual segmentation)."
accuracy, ROC (receiver operating characteristics) analysis
and faster training. Using the subsets of Lymphochip, this
paper then compared the performance of SVM against two
other well-known classifiers: multi-layer perceptron (MLP)
and linear discriminant analysis (LDA). Experimental results
show that SVM outperforms the other two classifiers."

Tencon 2005 Page 37

 1B-09
Session Title: Code Division Multiple Access 1
Chairperson: Yuu-Seng Lau
Day: Tuesday
Date: 22nd November 2005
Time: 11:20AM – 1:00PM
Room: M8

1B-09.1 1568964434 1B-09.4 1568965304

Optimization of UTRA SHO Algorithm Downlink Soft Handover Mechanism
Mahamod Ismail, Omar M. Eshanta, Norhayati Performance in WCDMA System
Ahmad, Muhammad Yusof Ibrahim Mahamod Ismail, Norazizah Mohd Aripin, Norhayati
"Since the emergence of W-CDMA technology as an air Ahmad, Muhammad Yusof Ibrahim
interface of 3G mobile communication systems, extensive "An in depth study of the soft handover effects on the
researches have been carried out to improve the downlink direction of WCDMA network is carried out. The
performance of these systems. Most of these works show link level study begins with the downlink interference
that the soft handover technique plays an important role in analysis and power allocations for each downlink dedicated
improving W-CDMA system performance. Several channels. Downlink soft handover gain is used in the
algorithms were developed to meet certain requirements of system level simulation as a main parameter to evaluate the
system such as capacity improvement, interference effects of cell selection and power control mechanisms on
reduction and traffic balancing [1][2]. Because of this the the downlink soft handover performance. Simulation shown
necessity to optimize these algorithms further increased. In that by using 2-way soft handover mechanism, the required
this paper we present a dynamic simulator using MATLAB dedicated transmission power up can be reduced up to 3
software to optimize the soft handover (SHO) parameters of dB. Besides that, the cell selection scheme based on
UTRA networks by taking the number of SHOs as an distance and Ec/I0 value gives better SHO gain (1.3 dB) to
optimization criterion. The simulations show that the the system with SHO overhead equals to 30%. Simulation
minimum number of SHOs can be obtained when the active also has shown that the system can benefit from the soft
set (AS) size, Add thereshold (Add_th) and Drop threshold handover process at 20% to 30% of the SHO overhead.
(Drop_th) were kept to 2, 2.5 and 3, respectively." However, power control mechanism does not give
significant impact to the SHO gain."
1B-09.2 1568964777
On the Performance of Walsh-LS Codes in CDMA 1B-09.5 1568963744
Systems Peter Conti Downlink Performance of a Finite Word Length
Upul Gunawardana, Ranjith Liyana-Pathirana Implemented Rake Receiver for 3G WCDMA
"Walsh-LS-CDMA (WLS-CDMA) systems is a novel Kumar V.
concept employing Loosely Synchronous (LS) codes and "In this paper, the performance degradation due to the finite
Walsh codes. The Walsh-LS spreading codes exhibit an word length implementation of the Rake receiver for 3G
Interference Free Window (IFW) in the auto-correlation and WCDMA is investigated. The performance in terms of bit
cross-correlation functions, which has the potential to error rate (BER) as a function of Eb/N0 in Rayleigh fading
suppress Inter Symbol Interference (ISI) and Multiple channels using Monte-Carlo computer simulations are
Access Interference (MAI), resulting in increased capacity carried out for varied word lengths. The performance results
compared to traditional CDMA systems. In this paper the are also compared with that of the floating-point
perspective of WLS-CDMA is explored. It is found that the simulations. The 8/6/4/2-bit word length implementations
forward-link has better performance than traditional CDMA are considered in the simulation runs in order to compare
systems, however in the reverse-link the Walsh-LS codes their effect on the performance."
lose orthogonality."

1B-09.3 1568964953
Blind Adaptive Multiuser Detection for DS-CDMA
Utilizing Sinusoidally-Distributed DSE-CMA
Jusak Jusak, Hanno Luehdorff, Zahir Hussain
The use of adaptive filtering to mitigate the multiple access
interference (MAI) in the direct-sequence code division
multiple access (DS-CDMA) systems is studied in this
paper. Blind adaptive algorithm based on our recently
proposed sinusoidally-distributed dithered signed-error
constant modulus algorithm (S-DSE-CMA) is introduced to
adapt the filter coefficients. This algorithm retains the
averaged transient behavior of the constant modulus
algorithm (CMA). We evaluate the performance of an
adaptive multi-user detector utilizing the S-DSE-CMA
scheme against the existed uniformly-distributed DSE-CMA
(U-DSE-CMA). It is shown that the new scheme converges
faster than the conventional scheme. Simulation results of
averaged mean-squared error (MSE) trajectories for
synchronous and asynchronous DS-CDMA are presented.

Tencon 2005 Page 38

 1B-10
Session Title: Power Systems Planning and Control 1
Chairperson: Ramesh Rayudu
Day: Tuesday
Date: 22nd November 2005
Time: 11:20AM – 1:00PM
Room: M10

1B-10.1 1568964636 1B-10.4 1568964908

A Design of the Ubiquitous-based Electric Power Damping Estimation In Highly Interconnected
Control Framework Power Systems
Mi Ra Kim Mark Glickman, Peter O'Shea, Gerard Ledwich
"Ubiquitous computing is the next generation computing "In a large interconnected power system, oscillating modes
environments with information & communication technology typically arise after disturbances. Estimation of the rate of
everywhere, for everyone, at all times. Information and rise or decay of these disturbances (i.e. of the damping
communication technology will be an integrated part of our factor) is important so that poorly damped modes which put
environments with integrated processors, sensors, and the system at risk can be identified. In the past both Fourier
actuators connected via high-speed networks. Ubiquitous- based and parametric techniques have been used to
based electric power control framework is a total method estimate the damping factors of the oscillating modes.
that measure electric power usage of each instruments. It Parametric methods tend to perform poorly in noise and
uses the RFID, sensor detecting and wireless network. The existing sliding window methods do not optimally account
purpose of this framework is that it would manage the for the kind of noise present in real operating environments.
requirement of electric power usage effectively, and inform This paper presents an improvement to existing Fourier
customer the electric power usage of accuracy." based methods for damping factor estimation. Simulations
and real data results are provided to support the claims
1B-10.2 1568964843 made in the paper. A comparison with parametric methods
is also made."
A Price-Based Unit Commitment Model
Considering Uncertainties 1B-10.5 1568965064
Rong-Ceng Leou A Novel Method on Substation Reactive Power
"In the regulated environment, unit commitment refers to Control with the Fuzzy Logic Method
optimizing generation resources to satisfy load demand at
least cost. But in a deregulated power market, generation
Yan XU, Qing LIU, Zengping WANG
companies have no obligation to satisfy customers "Based on analyzing the status and existing questions of
demands. Instead, they seek their maximum profits. This voltage and reactive power integrated control for
paper proposed a priced-based unit commitment algorithm substations, this paper presents a new voltage and reactive
for generation companies to maximize their profits. This power control system based on fuzzy logic and dynamic
algorithm considering uncertainties of energy prices was borderline applying fuzzy control theory. Simulation tests
solved by cooperation of the greedy algorithm and quadratic and practical running have proved that this approach can
program. A five-unit system was used to verify this reduce the adjust times of transformer tap and capacitor
algorithm." effectively and improve the control function."

1B-10.3 1568964907
Industrial Energy Audit using Data Mining Model
Web Application
Noor Maricar, MdNoah Jamal
"The excessive use of energy in industrial sectors
necessitates the decision maker to always question on how
the energy is being used efficiently. Energy used for air-
conditioning and lighting in a medium industry counts for
almost 60% of the total energy used. The small percentage
of energy use reduction relates to the lower product cost
and higher profit margins. Therefore, it is important to the
decision makers of an industry to have a proper method to
audit the building plant and to come up with the practical
actions needed in optimizing the use of energy, while at the
same time to improve the comfort and product quality. This
paper shows the data mining web application for energy
audit that can be used in a typical industrial site."

Tencon 2005 Page 39

 1B-11
Session Title: Future Challenges
Chairperson: Enn Vinnal
Day: Tuesday
Date: 22nd November 2005
Time: 11:20AM – 1:00PM
Room: M7

1B-11.1 1568963142 1B-11.4 1568965157

Intellectual Property - Challenges for the Future Neural Network Ensembles Based Approach for
David Webber Mineral Prospectivity Prediction
"Intellectual Property law is based on a balance between Vanaja Iyer, Chun Che Fung, Warick Brown, Kok-
the interests and needs of the public and those of creators. Wai Wong
When the legal systems that underpin intellectual property "In mining industry, accurate identification of new
no longer maintain the correct balance or even worse, geographic locations that are favourable for mineral
neglect it, then respect for those systems and intellectual exploration is very important. However, definitive prediction
property erodes. This paper discusses the current level of of such locations is not an easy task. In recent years, the
respect for intellectual property, and measures that could be use of neural networks ensemble approach to the
employed to addresses problems with the current system, classification problem has gained much attention. This
particulary for patents and copyright." paper discusses the results obtained from using different
neural network (NN) ensemble techniques for the mineral
1B-11.2 1568965296 prospectivtity prediction problem. The proposed model uses
The Detection of Curve-type Defects in the TFT- the Geographic Information Systems (GIS) data of the
location. The method is tested on the GIS data for the
LCD panels with Machine Vision Kalgoorlie region of Western Australia. The results obtained
WooSeob Kim, Yun Su Chung, Jong-Hwan Oh, Il are compared to some of the commonly known techniques:
Choi, Kil-Houm Park the majority combination rule, averaging technique,
weighted averaging method tuned by Genetic Algorithm
"The needs of automated inspection system are grown for (GA) and a newly proposed rule based method. The results
the Flat Panel Displays (FPD). However, it is so difficult to obtained using the different techniques are discussed."
detect the defects on the surface of FPD due to non-uniform
intensity distribution through the whole panel. Especially, 1B-11.5 1568965499
curve-type defects such as micro sized thread and scratch
are too difficult to detect with the conventional methods. In Establishment of Community Information
this paper, we propose a robust detection algorithm for Network in a Developing Nation
curve-type defects. The proposed algorithm is composed of
flattening, segmentation, and classification. The flattening Balaji Rajendran, Neelanarayanan, Ponraj
makes intensity distribution likely to be uniform. The Murugesan, Kailash Tandel
segmentation extracts candidates of curve-type defect and "In an era of information propagation through Internet, there
the classification selects real defects from candidates. The are many parts of the world, which have not benefited. This
experimental result verifies the performance of the kind of gap, known as 'digital divide' exists even in
proposed method." developed nations. The solution to the problem of bridging
this digital divide needs a platform for consuming, sharing,
1B-11.3 1568965302 and producing information in an easy-to-use and
The Exograsp delivers tactile and kinaesthetic understandable format, apart from the infrastructural
requirements. This paper discusses our solution in the form
information about virtual objects of a framework, ECKO. Our framework includes the
Barry Richardson, Mark Symmons, Dianne Wuillemin processes, methodologies, guidelines, and deployment
models apart from the software. The framework enables
We describe progress towards the provision of realistic
fruitful knowledge sharing among and across the
cutaneous information to accompany kinaesthetic input
communities and helps in creation of multi-dimensional
currently available in devices used to haptically explore
knowledge repositories. We also illustrate our phase-by-
virtual objects. A temperature change at the fingertips is
phase approach to the solution and it is generic enough to
made coincident with touching a virtual soft drink can.
be replicated across countries. The feedback obtained from
Pressure orthogonal to the skin surface contacting the
the stakeholders throughout the various phases is also
virtual surface can be incorporated in a similar way.
discussed and is very much encouraging."

Tencon 2005 Page 40

 1B-12
Session Title: Array and Space-Time Signal Processing 1
Chairperson: Zahir Hussain
Day: Tuesday
Date: 22nd November 2005
Time: 11:20AM – 1:00PM
Room: M9

1B-12.1 1568964831 our newly considered receiver called cooperative maximum

likelihood detector while the others are maximum likelihood
Partial Discharges Multi-Targets Localization in detection in [12], Alamoutis receiver, zero-forcing detection
Power Transformers Based On Array Signal and decision-feedback detection. Our results show that all
Processing detection methods, except cooperative maximum likelihood
detection, achieve nearly the same BER performance over
R.C LUO
time-varying channels, unlike the results obtained in [12] by
“A new partial discharges multi-targets localization method assuming perfect channel estimation. In many cases,
is presented in this paper, which is based on array signal cooperative maximum likelihood detection performs the
processing. Two ultrasonic sensor arrays are located on the best by about 5dB due to the diversity gain. Also, the
inner wall of the transformer tank and detect the ultrasonic influence of timevarying channels results in the error
signal emitted by the partial discharge sources. Spatial flooring effect."
spectrum estimation technique is used in analyzing the
signal characteristic information (number of signals, 1B-12.4 1568964838
direction-of-arrivals et. al) and the number of partial
discharge signal estimation, direction-of-arrival estimation Performance Analysis of Multiple-Relay Decode-
and distance estimation are accomplished. The and-Forward Cooperation System
performance of localizations is given by computer Jun Zhang, Tat Lok
simulations. It shows that the method processes with high
resolution and good estimated accuracy and can effectively "In this paper, we analyze the impact of the detection error
estimate the location parameter of multiple partial discharge at the relay node to the performance of the decodeand-
sources in transformer." forward cooperation system. We apply a simple detector,
for which no channel state information between source
1B-12.2 1568964834 node and relay node is available at the destination, and
maximum ratio combining is performed without considering
Space-Time Encoded Secure Chaos whether the forwarded symbol from the relay is incorrect.
Communications with Transmit Beamforming With such an assumption, we analyze the error
performance of the system with multiple relay nodes, and
Lau Yuu Seng, Kevin Lin, Zahir Hussain
demonstrate the resulting error floor in this system. To
"It is shown that the use of chaos shift keying (CSK) with a compensate for the detection error at the relay node, we
transmit diversity technique such as beamforming and propose a relay selection protocol, which can lower the
space-time coding can provide a secure communication link error floor and nearly achieve full diversity order."
with improvement in the system error performance.
Spreading sequences that are used to scramble data in 1B-12.5 1568965130
spread spectrum (SS) systems can be generated using a
single mathematical relationship of a chaotic generator. Error Performance Analysis of Clipped Alamouti
These chaotic sequences are very difficult to predict over Space-Time Coded OFDM Systems
long-term chaotic pattern unless the exact initial condition of Himal Suraweera, Jean Armstrong
the chaotic generator is known, providing security. On the
other hand, beamforming and orthogonal space-time block “In this paper we investigate the performance of clipped
coding (OSTBC) transmit diversity techniques are known to Alamouti space-time coded (STC) orthogonal frequency
provide optimal transmitting structures for communication division multiplexing (OFDM) systems. The high peaks in
systems, especially if combined. Based on the signal angle- STC-OFDM signals are clipped to limit the peak-to-average
of-arrival (AoA) estimates, the channel correlation matrix power ratio (PAPR). In STC-OFDM and OFDM, clipping can
can be constructed, and it is shown that signal transmission achieve significant PAPR reduction while maintaining a low
of OSTBC codes in the eigen-modes of this matrix gives an symbol error rate (SER). Previous studies on OFDM have
effective array weighting gain which improves system error concluded that the effect of channel fading on clipping noise
performance without sacrificing any diversity and coding must be considered for an accurate error performance
gain. A performance study of CSK with beamforming and analysis. We extend this further to incorporate STC-OFDM
space-time encoded is carried out in this paper." systems and jointly study the impact of channel fading and
receiver combining on clipping noise. At the STC-OFDM
1B-12.3 1568964837 receiver, effective clipping noise governing the signal-to-
noise ratio has different characteristics compared to the
Detection Schemes for Distributed Space-Time same for OFDM. Simulations indicate that STC-OFDM is
Block Coding in Time-Varying Wireless more sensitive to clipping effects than OFDM. The
Cooperative Systems theoretical and simulated SER are also compared for
clipped STC-OFDM."
Ho Ting Cheng, Tat Lok
“In this paper, we investigate the influence of time varying
channels on the bit-error-rate (BER) performance for
distributed space-time block coded system. We make use
of the pilot symbol assisted modulation (PSAM) to estimate
the time varying channel coefficients for assessing the
performance of different reception methods, one of which is

Tencon 2005 Page 41

 1C-01
Session Title: Traffic and Resource Management 2
Chairperson: Irena Atov
Day: Tuesday
Date: 22nd November 2005
Time: 2:00PM – 3:40PM
Room: P2

1C-01.1 1568964909 calls to monopolise the system resources at the expense of

voice call blocking. The PE scheme remedies this by
Using Rate Balancing to Improve Throughput in a providing priority to voice calls. However, in demand
Wireless Multi-hop Chain Topology scenarios where data is the dominant traffic, voice calls can
Binh Ngo, Steven Gordon, Ha Duong receive unrealistically good performance at the expense of
increased delay for data calls. Overall, PE results in a more
"A well-known problem in wireless multi-hop communication efficient use of the system resources, especially with high
is the significant degradation of throughput when the demand."
number of hops increases. Previous research shows that in
a chain topology, nodes using IEEE 802.11 RTS/CTS can 1C-01.4 1568964822
at best achieve 1/4, but in practice only 1/7, much smaller
than the ideal 1/3 of the throughput of one-hop Traffic Measurement for Large Internet Flows
transmission. In order to improve throughput performance in Using Multi-stage Landmark LRU Cache
multi-hop cases, we introduce a rate balance scheme and
tune the control parameters of sensing range at physical
Lichang Che, Bin Qiu
layer and contention window at MAC layer. Our simulation "The scalability issue of the Internet management has
results show that the throughput ratio close to 1/4 is motivated recent research on partial flow state based traffic
achievable when the number of hops is as high as 12; measurement and management. Based on our previous
better ratios of 1/3.4 and 1/3.6 are achieved when number Landmark LRU proposal for long-term fast flow
of hops is 5 and 8, respectively." detection\cite{che-05-llru}, this paper proposes a novel
traffic measurement scheme with partial flow state, Multi-
1C-01.2 1568965144 stage Landmark LRU (ML-LRU) cache. Inherited the merit
of the Landmark LRU cache in reducing the possibility of
Using Reinforcement Learning for Call large flows being flushed out of the cache, the ML-LRU
Admission Control in Cellular Environments locks large flows reliably in the cache for traffic
featuring Self-Similar Traffic measurement. Simulations driven by real Internet traces
prove that the ML-LRU scheme uses only one third of the
Nimrod Lilith, Kutluyil Dogancay
memory space required by the sample-and-hold \cite{estan-
This paper details reinforcement learning architectures that 02-measure} scheme and meanwhile improves
efficiently provide the functions of dynamic channel measurement performance notably."
allocation (DCA) and call admission control (CAC) for
cellular telecommunications environments featuring both 1C-01.5 1568965354
voice traffic and self-similar data traffic. These solutions are
able to be implemented in a distributed manner using only Effective Verification of Wavelength Routing
localised environment information and without the need for Function by Wavelength Transfer matrix Method
any off-line training period. The performance of these Kimio Oguchi, Kohei Okada
reinforcement learning solutions are thoroughly examined
via computer simulations and are shown to produce "The wavelength transfer matrix is a powerful approach to
superior results in terms of both revenue raised and handoff design and verify the wavelength routing needed for
blocking probability. complicated wavelength routing networks. This paper first
describes the wavelength transfer matrix, the basic form
1C-01.3 1568965185 and some extensions. The WDM PON networks proposed
to date are analyzed in terms of wavelength by using the
Hybrid Delay/Blocking Wireless Systems with wavelength transfer matrix. The results show the method is
Greedy Resource Allocation effective and practical for designing complicated networks."
Paul Fitzpatrick, Milosh Ivanovich, Philip Bickerdike
"This paper presents an analysis of a hybrid delay /blocking
system model with wide applicability in wireless networks.
Real time voice traffic operates under a blocked calls
cleared regime while non-real-time data is delayable. These
traffic types compete for a fixed number of servers under
two schemes: (i) first come first served (FCFS) and (ii) pre-
emption (PE) of ongoing data calls by new voice calls. Each
voice call uses one server, while data calls are served in a
First-In First-Out (FIFO) queue with capacity equal to all
servers not occupied by voice. Voice and data calls are
modelled as Poisson arrivals with exponential holding
times, and the resulting continuous-time Markov process is
analysed using a stochastic simulation of the state space.
The results presented demonstrate trade-offs between the
performance of real-time voice and delayable data traffic, as
well as the differences between this hybrid system and
conventional systems based solely on blocking or delay.
We show that the use of the FCFS scheme causes data

Tencon 2005 Page 42

 1C-02
Session Title: Internet Applications 1
Chairperson: Greg Egan
Day: Tuesday
Date: 22nd November 2005
Time: 2:00PM – 3:40PM
Room: M1

1C-02.1 1568964547 1C-02.3 1568964883

Towards Robust Internet Applications: Self- Intelligent Customer Relationship Management
Organizing Overlays On The Web
John Risson, Tim Moors Kok-Wai Wong, Chun Che Fung, Xiong Xiao, Kit Po
"Self-organizing overlays are set to make internet Wong
applications more robust. They can improve system "In recent years, Customer Relationship Management using
dependability, evolve smoothly from small configurations, web personalisation initiatives have gained much attention.
and scale to massive configurations. Application case The most important strategy of web personalisation is to
studies are presented to illustrate weaknesses of current provide the customers with correct information or services
architectures and to show how overlays can improve based on the knowledge about the customers preferences.
robustness-cost tradeoffs. The case studies include email, With the help of data mining technologies, the above
next generation voice and mobile networks based on the strategy can be implemented. Computational intelligence
Session Initiation Protocol (SIP), network access protocols technologies are investigated in this paper to provide
like Radius and Diameter, and publish/subscribe interaction through web personalisation. This paper
middleware. A program of work is outlined to transition proposed two algorithms for the personalisation of the
overlay technology from the laboratory to industry." online shopping websites. It uses the Radial Basis Function
(RBF) neural network. The algorithms first model the
1C-02.2 1568964864 customers preferences as a complex nonlinear function. It
personalises the information presented to customers based
Evaluating the impact of DNS and HTTP session on their preferences. The second is the preference learning
characteristics on consumer ISP web traffic algorithm. It learns the customers preferences implicitly
Jason But, Urs Keller, Grenville Armitage from the customers behaviours by using a RBF neural
network."
"Web caches are generally considered useful because they
reduce replication of network traffic flowing from original 1C-02.4 1568964922
content sources. In this paper we experimentally
characterise the network and transport layer consequences Context Based Riskiness Assessment
of web caching in the consumer ISP context. We
instrumented a small number of Australian, broadband-
Omar Hussain, Elizabeth Chang, Farookh Khadeer
attached homes to collect round-trip time (RTT) and hop Hussain, Tharam Dillon, Ben Soh
count statistics for their HTTP/TCP sessions, and collect "In almost every interaction the trusting peer might fear
DNS lookup statistics associated with each HTTP about the likelihood of the loss in the resources involved
exchange. We estimated the impact of DNS lookup delays during the transaction. This likelihood of the loss in the
on overall HTTP session times, and use our RTT and hop resources is termed as Risk in the transaction. Hence
count statistics to show that consumer ISPs would benefit analyzing the Risk involved in a transaction is important to
greatly from local caching, particularly in Australia where decide whether to proceed with the transaction or not. If a
speed of light delays have a large impact on session times trusting peer is unfamiliar with a trusted peer and has not
when retrieving international content." interacted with it before in a specific context, then it will ask
for recommendations from other peer in order to determine
the trusted peers Riskiness value or reputation. In this
paper we discuss the process of asking recommendations
from other peers in a specific context and assimilating those
recommendations according to its criteria of the interaction
in order to determine the correct Riskiness value of the
trusted peer."

1C-02.5 1568965043
The Design of China Tourism E-Business
Platform Based on J2EE Technology
Wensheng Gwsice, Wensheng Guo, Yixin Yin,
Xuyan Tu
This paper presented the problems met during the
development of an e-business platform using J2EE
technique and gave the solutions to solve these problems.
We discussed the detailed design and implementation
issues in developing a China Tourism e-business platform.

Tencon 2005 Page 43

 1C-03
Session Title: Image Processing 2
Chairperson: Henry Wu
Day: Tuesday
Date: 22nd November 2005
Time: 2:00PM – 3:40PM
Room: P1

1C-03.1 1568964826 1C-03.4 1568964988

A motion-based approach for temporal texture A Nonparametric Discriminant Approach in
synthesis Resolving Complex Multi-class Query for
Ashfaqur Rahman, Manzur Murshed Content-based Image Retrieval
In this paper we propose a motion based approach for Kien Ping Chung, Chun Che Fung
temporal texture synthesis. Temporal textures are natural "Content-based image retrieval (CBIR) systems have drawn
phenomenon characterized by their distinctive motion intense interest from many researchers in recent years.
patterns. Synthesis of these textures is thus considered as Over this period, certain degree of success has been
regeneration of motion pattern that has identical motion achieved in domain-oriented systems for applications such
distribution of a source texture. In this paper we propose a as facial recognition and medical diagnosis. However, the
synthesis technique where new textures are generated by machine learning techniques used in these systems mostly
computing their movement pattern from a known motion assume that all the targeted images belong to a single
distribution followed by generation of image frames. group. Thus, most of the research efforts so far have been
Experimental results demonstrate the ability of the trying to search for one or a combination of global image
proposed technique by producing visually promising features that can be used to differentiate the targeted
temporal textures. images from the rest. This is not the case for a generic
image database. Quite often, images that are similar
1C-03.2 1568965098 semantically may be completely different with the visual
context. In this paper, the authors propose a local grouping
The Development of a Novel Image Quality Metric strategy together with a multiple Gaussian distributions
and a Synthetic Colour Test Image for Objective distance ranking approach in an attempt to address the
Quality Assessment of Digital Codecs retrieval and ranking of images that belong to multiple
Amal Punchihewa, Donald Bailey, Bob Hodgson disjoint groups."

"This paper proposes a novel image quality metric for 1C-03.5 1568965013
colour codecs and a novel colour test image. The degree of
colour error (bleeding) due to image compression is Subregion Technique with Optimal Local
evaluated with a special colour test image designed to Parameter Sets for Image Deconvolution
emphasise colour artefacts. The known reference
approach, which is used in this research, offers swift and
Karn Patanukhom, Akinori Nishihara
accurate measurements. The proposed test image replaces "A subregion technique for image deconvolution is
the full field static colour bar test signal used in analogue proposed. This technique improves restoration results of
television and video engineering. The proposed test image conventional deconvolution methods by separately restoring
can be used to evaluate the performance of digital video each local region of an image with different restoration
and television broadcasting facilities. The objective quality processes. Optimal local restoration parameter set is
measure can also be used in the codec development designed for each restoration process to optimize a
process and in benchmarking codec performance. The restoration error in its corresponding local region. An
efficacy of the new metric and test image is assessed using iterative algorithm to estimate the optimal parameter sets
a JPEG codec at a range of compression levels." for subregion scheme is introduced. Instead of using an
original pair, which is a pair of original and degraded
1C-03.3 1568964868 images, to find a characteristic between the restoration
error and the parameter set, the proposed algorithm uses a
A Sampling Switch Design for Liquid Crystal pseudo pair, which is a pair of an estimated original image
Displays and its degraded version having the same degradation
Shingo Takahashi, Shuji Tsukiyama, Masanori condition as the original pair. Simulation results are
presented to demonstrate the promising performance of the
Hashimoto, Isao Shirakawa
subregion technique."
“In the design of a column driver in an active matrix LCD,
the main issue is how to keep the admissible allowance of
the RPV (Ratio of Pixel voltage to Video voltage) of each
pixel within a specified range, which determines the
transmitted luminance of the pixel. This constraint on RPV
is analyzed in terms of a minimal number of parameters
associated with the sampling switch and the sampling
pulse, and then a design scheme is described dedicatedly
for the sampling switch. Experimental results show that an
optimal sampling switch can be attained, which gives rise to
a column driver with almost 50% less power consumption
than the one by manual design."

Tencon 2005 Page 44

 1C-04
Session Title: IT Applications 2
Chairperson: Horace King
Day: Tuesday
Date: 22nd November 2005
Time: 2:00PM – 3:40PM
Room: P3

1C-04.1 1568965148 1C-04.4 1568965228

Applying Machine Learning Techniques to Collaborative Editing using an XML Protocol
improve Linux Process Scheduling Stephen Davis, Ian Burnett
Kishore Kumar Pusukuri, Atul Negi "XML is a popular approach to interoperable exchange of
"In this work we use Machine Learning (ML) techniques to data between a wide range of devices. This paper explores
learn the CPU time-slice utilization behavior of known the use of the Remote XML Exchange Protocol as a
programs in a Linux system. Learning is done by an mechanism to provide efficient interaction with complex
analysis of certain static and dynamic attributes of the XML documents to users with limited complexity devices
processes while they are being run. Our objective was to and/or limited bandwidth connections. The interactive
discover the most important static and dynamic attributes of mechanisms provided by the protocol allow users to
the processes that can help best in prediction of CPU burst navigate, edit and download XML even when delivery of the
times which minimize the process TaT (Turn-around-Time). full XML document is impossible. The paper examines the
In our experimentation we modify the Linux Kernel use of the protocol to enable multiple users to
scheduler (version 2.4.20-8) to allow scheduling with collaboratively edit remote XML documents. Further, the
customized time slices. The ``Waikato Environment for paper explores the combination of the protocol,
Knowledge Analysis"" (Weka), an open source machine- collaborative editing and re-cently released Word
learning tool is used to find the most suitable ML method to processor/Office suite XML schema formats."
characterize our programs. We experimentally find that the
C4.5 Decision Tree algorithm most effectively solved the 1C-04.5 1568965231
problem. We find that predictive scheduling could reduce
TaT in the range of $1.4\%$ to $5.8\%$. This was due to a
The Performance of Reliable Server Pooling
reduction in the number of context switches needed to Systems in Different Server Capacity Scenarios
complete the process execution. We find our result Thomas Dreibholz, Erwin Rathgeb
interesting in the context that generally operating systems
presently never make use of a program's previous "Reliable Server Pooling (RSerPool) is a protocol
execution history in their scheduling behavior." framework for server pool management and session
failover, currently under standardization by the IETF
1C-04.2 1568965150 RSerPool WG. While the basic ideas of RSerPool are not
new, their combination into one architecture is. Some
Introducing NTS: A Classification System for research into the performance of RSerPool for certain
Knowledge Library Applications specific applications has been made, but a detailed,
application-independent sensitivity analysis of the system
Eric Rayner, Ian Piper, Martin Bunder parameters is still missing. The goal of this paper is a to
"This paper identifies and defines the Knowledge Library systematically investigate RSerPool's load distribution
Application (KLA) class. KLAs are an important class of behaviour on changes of workload and system parameters
information resource/knowledge management applications. to give basic guidelines on designing efficient RSerPool
It will be clear from the definition that KLAs are highly systems. In this paper, we especially lay our focus on
relevant to both the commercial and academic worlds and scenarios of server pools consisting of servers with unequal
existing techniques such as keyword and tree classification, capacities."
On-Line Analytical Processing (OLAP), and Formal Concept
Analysis (FCA) do not provide key KLA features. N-Tree-
Space (NTS), a technique detailed in this paper, enables
the provision of these features and should facilitate the
development of a range of KLAs that offer significant
advantages over currently available software tools."

Tencon 2005 Page 45

 1C-05
Session Title: Modulation and Coding 2
Chairperson: Zahir Hussain
Day: Tuesday
Date: 22nd November 2005
Time: 2:00PM – 3:40PM
Room: M2

1C-05.1 1568964894 1C-05.4 1568965251

Performance of the Time-Delay Digital Tanlock Analog Baseband Chain in a 0.18 µm
Loop as PM Demodulator in Additive Gaussian Standard Digital CMOS Technology for
Noise IEEE802.15.3a (UWB) Receiver
Zahir Hussain Jeongwook Koh, Hanseung Lee, Jung-Eun Lee,
"In a previous work we proposed a phase-lock structure Choong-Yul Cha, Hyunsu Chae, Eun-Chul Park
called the time-delay digital tanlock loop (TDTL). This digital Today UWB (Ultra-wideband) communication is gaining a
phase-locked loop (DPLL) performs nonuniform sampling great attention from consumer electronic-industry and is
and utilizes a constant time-delay unit instead of the expected to be adopted for many WPAN (wireless personal
constant 90-degrees phase-shifter used in conventional area network) applications. UWB (Ultrawideband)
tanlock structures. The TDTL reduces the complexity of communication standard is based the form of direct
implementation and avoids many of the practical problems sequence or OFDM (orthogonal frequency division
associated with the analog or digital Hilbert transformer, multiplexing) for high-date-rate application like IEEE
while it preserves the most important features of the 802.15.3a [1]. This paper demonstrates a successful
conventional DTL (CDTL). In this paper we investigate the implementation and experimental verification of a CMOS
performance of the TDTL in demodulating phase- and analog baseband chain - variable gain amplifier and low
frequency-modulated signals in the presence of additive pass filter - for a UWB receiver in a 0.18µm CMOS
Gaussian noise." process. The experiments features 40 dB linear scaled
output gain and a total current of 30 mA.
1C-05.2 1568964958
Long-haul ASK and DPSK Optical Fibre 1C-05.5 1568965309
Transmission Systems: Simulink Modeling and Stage-Hopping Memory Management Technique
Experimental Demonstration Test-Beds for Viterbi Decoders
Thanh Huynh, Le Binh, Quoc Huy Lam, Dung Tran Guan-Henry Lin, Tsern-Huei Lee
"We present the MatlabTM Simulink model and its "A new memory management technique for survivor paths
corroboration with experimental transmission of amplitude in Viterbi decoder is proposed in this paper. It is based on a
and differential phase modulation incorporating RZ, NRZ, central concept that states reserved for merely certain
carrier suppressed RZ formats over 320 km long-haul stages will be enough to decode out the whole sequence.
optical fibre transmission systems. Dispersion tolerance of Thus, traceback could be complete faster since it could be
the transmission systems under non-compensation and realized block-by-block instead of stage-by- stage.
compensated transmission fibres is given. Simulated and Furthermore, the central concept leads to a new memory
measured eye diagrams are reported." management method that records states only for those
necessary stages. Thus, compared with other methods, the
1C-05.3 1568965244 major drawback of TBM- decoding latency could be greatly
reduced approaching the extent of REA and up to 45% of
A Coherent Detection-based Symbol Detector the memory size could be saved in hardware
algorithm for 2.45GHz LR-WPAN receiver implementation."
Joo-Hyun Do, Jung-Su Han, Usang Lee, Sang-Hyun
Min, Hyung-Jin Choi
"In IEEE 802.15.4 LR-WPAN(Low-Rate Wireless Personal
Area Network) specification, the frequency offset of 80 ppm
on 2.45GHz band is recommended for low-cost
implementation. For robustness to the frequency offset,
non-coherent detection-based symbol detector algorithm is
typically applied in the LR-WPAN receiver modem.
However, the noncoherent symbol detector has more
performance degradation by squaring loss of I/Q squaring
operation than the coherent detector. In this paper, we
propose a coherent detection-based symbol detector
algorithm with frequency offset compensation that has
better performance and lower complexity than the
noncoherent detector algorithm without additional
complexity due to the frequency offset compensation."

Tencon 2005 Page 46

 1C-06
Session Title: Time-Frequency Analysis and Wavelets 2
Chairperson: Peter O’Shea
Day: Tuesday
Date: 22nd November 2005
Time: 2:00PM – 3:40PM
Room: P2

1C-06.1 1568964793 1C-06.3 1568964844

A Wavelet Based Feature Extraction For Voice A Wavelet Packet Modulation Method for Satellite
Lock Systems Communications
Pranav Kumar CHERUPALLI, Jaya Sankar Hideto Sakakibara, Eiji Okamoto, Yasunori Iwanami
KOTTAREDDY GARI "Wavelet Packet Modulation (WPM) was proposed as one
"Abstract - In an approach to provide more secure, fast and of the multicarrier transmission methods like OFDM. Since it
efficient speaker verification system (Voice Lock System), is a multicarrier transmission method, it has a high peak to
this paper reviews the fast wavelet feature extraction averaged power ratio (PAPR). For improving PAPR of
technique and wavelet packet technique for speech OFDM, the Single Carrier OFDM (SC-OFDM) method was
recognition systems. One of the biggest advantages of the proposed and it is expected that the principle of SC-OFDM
Wavelet Transform over the Short Time Fourier can be used in WPM. In this paper, we consider applying
Transform(STFT) is its capability to process nonstationary the WPM to satellite communications. The BER
signals. As the prosodic features of speech signals in the performance of WPM and OFDM in a fading channel is
voice lock systems play an important role, these features compared, and it is shown that the performance of WPM is
have to be extracted with great accuracy, which are better than OFDM. Then, we propose the Single Carrier
efficiently extracted using the Wavelet Packets which WPM (SC-WPM) method that applies the principle of SC-
provide the local time-spectral description. The adaptive OFDM to WPM to remedy the PAPR, and show that the
frame synchronization method intelligently classifies these performance is improved substantially."
wavelet packets such that all the prosodic features are
extracted efficiently. The paper also talks about the 1C-06.4 1568964885
problems with the HMM systems in view of the security of
the ASR systems."
Viability of a wavelet-based multicarrier
modulation system
1C-06.2 1568964841 Quoc Tuan Nguyen, Dinh Nguyen
An algorithm for estimating time-varying Doppler "The authors consider the viability of a wavelet-based
at low SNR multicarrier modulation (MCM) scheme as an alternative to
the current FFT-based system used in an asymmetrical
Maree Farquharson, Peter O'Shea, Gerard Ledwich digital subscriber line (ADSL). Current work tends to unfairly
"This paper presents an algorithm for estimating the time- compare the performance of wavelet-based MCM to that of
varying Doppler return signal from a target which has a the conventional MCM. Biorthogonal wavelet bases have a
time-varying acceleration, with this changing acceleration quite different structure from the classical FFT-based
being modeled as a linear function of time. The method harmonic structure. Within each subband in the proposed
involves modeling the return signal as the sum of a number wavelet-based MCM system, we have essentially a TDM
of constant amplitude cubic phase signals embedded in arrangement. Thus a wavelet-based MCM system is
noise. An algorithm is presented for estimating the phase practically a combination of orthogonal TDM (between
parameters and subsequently the time-varying Doppler translates) and orthogonal scale division multiplex (OSDM).
(under the assumption that the Doppler shift is small Unless this particular structure of a future wavelet-based
compared with the center frequency of the irradiation). MCM system can be better understood and exploited for
Simulations are presented to confirm the effectiveness of appropriate channel equalisation, channel loading and
this algorithm." signal demodulation, it is premature and will not be
meaningful to compare the performance of a wavelet-based
MCM system to that of the current FFT-based system."

Tencon 2005 Page 47

 1C-07
Session Title: Electronic Devices 2
Chairperson: Amin Sadik
Day: Tuesday
Date: 22nd November 2005
Time: 2:00PM – 3:40PM
Room: M4

1C-07.1 1568964957 1C-07.3 1568965014

Analysis of Nonlinear Propagation and Wave Diode fabricated by layer by layer deposition of
Mixing of Picosecond Pulses in Semiconductor semiconductor nanoparticles
Optical Amplifiers Joydeep Dutta, Chanchana Thanachayanont
Narottam Das, Nemai Karmakar, Yasuhiro We report the development of a modified polyelectrolyte
Yamayoshi, Hitoshi Kawaguchi deposition technique for building multilayer thin films. The
"We have analyzed the nonlinear propagation quantum dots of zinc sulphide nanoparticles doped with
characteristics and wave mixing of picosecond pulses in manganese were synthesized by co-precipitation
semiconductor optical amplifiers (SOAs) by the finite- techniques and the nanoparticles were found to consist of
difference beam propagation method (FD-BPM). From the crystallites that were estimated to be around 2.2 nm. The
simulation results, it was found in the output waveform that modified polyelectrolyte deposition technique has been
the pulses peak positions are shifted to the leading edge. used for the growth of multifunctional nanoparticulate ZnS
The output spectra are red-shifted and the dips are thin films and schottky diodes. The Schottky junction device
observed at the higher frequency side in the frequency exhibit diode-like behavior and this technique shows
spectra. We have analyzed all-optical demultiplexing promise for large area construction of diodes using wet-
(DEMUX) based on four-wave mixing (FWM) in SOAs. It is chemical synthesis of semiconductor nanoparticles.
clear from the simulation results that the SOAs with wider
gain bandwidth are required for the faster DEMUX 1C-07.4 1568965024
operation. We have simulated pattern effects in the FWM Low-voltage CMOS instrumentation amplifier for
signal. We have also obtained DEMUX from the time-
multiplexed signals by repetitive pump pulses. FWM signal
piezoresistive transducer
intensity decreases with the strong energy pump pulses. Suttisak Sangtong, Apinunt Thanachayanont,
However, there is no pattern effect due to gain saturation Chanchana Thanachayanont
because the pump pulses are injected continuously."
A low-voltage CMOS instrumentation amplifier for
1C-07.2 1568964964 piezoresistive transducer is proposed. An input current
sensing configuration is used to allow the use of a single
Use of High-Tc Superconducting Tape RF Coils piezoresistor and low voltage operation. The amplifier is
for Non-invasive Fiber Tracking on Diffusion realized using a 0.35-um CMOS technology and operates
Tensor MRI application under a single 1.5-V power supply voltage. Simulation
results showed that the amplifier can achieve the maximum
In-Tsang Lin, Chang-Wei Hsieh, Li-Wei Kuo, Hong- CMRR of 139 dB while dissipating 489 uW.
Chang Yang, Herng-Er Horng, Wei-Hao Chang
"High-temperature superconducting (HTS) coil is one of the 1C-07.5 1568965247
best ways to increase the signal-to-noise ratio (SNR). A Novel Monostable Multivibrator Circuit
Bi2Sr2Ca2Cu3Ox (Bi-2223) tapes were suitable to use
because of the easier fabrications and lower cost. In this Panwit Tuwanut, Jeerasuda Koseeyaporn, Paramote
study, we built HTS Bi-2223 tape coils and demonstrated Wardkein
that the SNR of using the HTS tape coil was 3 or 4 folds "This article represents a novel monostable multivibrator
higher than that of the traditional copper coil for a rat brain circuit in which width and height of the pulse output can be
MR study. Acquisition time of MR diffusion tensor imaging independent electronically adjusted. In addition, the pulse
(DTI) can be reduced by factor of 9 for the same signal-to- width of output is not restricted to be wider or narrower than
noise. Accuracy of fiber tracking using DTI is also that of input pulse as conventionally employed. Namely,
significantly improved by a factor of 2.5 or so using HTS there is no restriction on the width of input and output pulse,
coil. In summary, with this HTSC system, a 3T MR system the circuits implementation is thus more convenient. The
could reach the high signal-to-noise of 12 T MR system with proposed circuit is composed of three OTAs, two resistors
the advantage of less T2 shortening effects at high field. and one capacitor, basically constructed a simple pulse
Currents researches are focused on brain connectivity and generator (Schmitt-trigger circuit). Therewith the
fMRI studies." compactness of the circuit, it is thus suitable to be
fabricated for integrated circuit. The results of circuit
simulation is accordance well with theoretical analysis."

Tencon 2005 Page 48

 1C-08
Session Title: Biomedical Engineering 2
Chairperson: Ashkan Sami
Day: Tuesday
Date: 22nd November 2005
Time: 2:00PM – 3:40PM
Room: M6

1C-08.1 1568965199 orientation. Performance of the proposed technique was

examined experimentally and promising results were
FES as Biofeedback for an EMG Controlled obtained."
Prosthetic Hand
Alejandro Hernandez Arieta, Hiroshi Yokoi, Tamio 1C-08.4 1568965421
Arai, Wenwei Yu A Study of Power Asymmetry Ratio of Single
"The importance of providing with biofeedback when Trial Electroencephalogram During Finger
interacting with man-machine interfaces has to be Movement
considered when developing new applications, in particular,
with externally powered prosthetic devices. In this study we
Kok-Meng Ong, Raveendran Paramesran
look to develop a tactile biofeedback system for this "This paper presents a study on Power Asymmetry Ratio of
purpose using electrical stimulation. In this paper we study the Single Trial Electroencephalogram (EEG) during the
the ability of the human body to recognize different patterns preparation of left or right finger movement, or more
of stimulation using two methods: surface and interferential precisely self-paced tapping. The objective is to predict
current stimulation and their influence over the EMG which finger is going to be used to press a key in the
acquisition process. The goal is to develop an appropriate computer keyboard. The change in EEG signal is more
feedback source to the human body to be used along with pronounced in the contralateral hemisphere. Therefore, the
an EMG controlled prosthetic hand." use of Power Band Asymmetry Ratio will enhance the
identification of left or right finger tapping. Parametric
1C-08.2 1568965214 spectral estimation, in this case autoregressive (AR)
spectral analysis with Burg method is used to obtain the
Bluetooth-enabled ECG Monitoring System Power Spectral Density (PSD) from 8Hz to 30Hz. The
Desmond Kho, Rosli Besar, Yeong Shun Tan, Kok proposed method selects the most critical homologous
How Tee, Kim Choo Ong channels in the classification of the task. It is discovered
that the homologous pair of CP3 and CP4 is the most
"Heart disease continues to be the leading cause of important channel-pair that gave a classification accuracy of
mortality in Malaysia in an era of industrialization, changing seventy five percent."
lifestyles and dietary habits. The mortality rate reported in
government hospitals shows an increase in death due to 1C-08.5 1568965532
heart diseases, and this trend is expected to continue into
the next decades. Since heart disease is increasingly Decision Tree Construction for Genetics
affecting human lifestyle, rehabilitation and practical devices Applications Based on Association Rules
are being carried out and created in order to reduce the
disability from heart diseases. This paper proposed a
Ashkan Sami, Makoto Takahashi
wireless patient monitoring system which integrates "A decision tree is an effective means of data classification
Bluetooth technology. In this paper, a detailed description of from which rules can be both expressive and precise.
constructing a 2-lead electrocardiogram (ECG) sensor, However decision tree is only applicable in the applications
transmitting the ECG data acquired from the ECG sensor that the data is expressed with attribute-value pairs. Since
via the Bluetooth wireless link, and receiving the ECG data genetic data are not attribute-pairs, the only method that we
at the PC is illustrated. The acquired ECG data is know of to make decision-tree for them is based on a
processed, manipulated and constructed as an ECG greedy graph-based data mining algorithm known as DT-
waveform. Lastly, the ECG waveform is displayed on the GBI. Though to its greedy nature, some of the important
personal computer (PC) screen. The results show that rules may be missed. Even though recently some attempts
implementation of wireless technology in the existing ECG to make the algorithm complete has been presented, the
monitoring system eliminates the physical constraints computational complexity of algorithm increased so much
imposed by hard-wired link and allows users to conduct that is not appropriate for practical purposes. In this paper
own check up at anytime anywhere." we present an approach to make decision tree for DNA
based data, which basically uses regular association-rule
1C-08.3 1568965392 algorithms. Thus it has computational complexity is much
more tractable. Our contribution in this paper is to convert
Human Motion Representation Using DNA based problems to regular data mining methods and
Eigenspaces by this conversion we present a method that can be applied
Joo Kooi Tan, Satoshi Houman, Seiji Ishikawa to all classes of classification based on association rules."

"Recognition of human motions using their 2-D appearance

images has various applications. An eigenspace method is
employed in this paper for representing and recognizing
human motions. An eigenspace is created from video
images taken by multiple cameras that surround a human in
motion. The image streams of the motion compose several
curved lines in the eigenspace that are closely situated with
each other. The motion is described by this set of lines,
which ultimately compose a curved surface. It is used for
recognizing a human motion observed from an arbitrary

Tencon 2005 Page 49

 1C-09
Session Title: Code Division Multiple Access 2
Chairperson: Yuu-Seng Lau
Day: Tuesday
Date: 22nd November 2005
Time: 2:00PM – 3:40PM
Room: M8

1C-09.1 1568965117 1C-09.3 1568965268

Performance analysis and comparison of trellis Use of Turbo Coding for Embedded
coded STTD and TSTD schemes for CDMA Transmission Scheme in Optical CDMA
system Tomoyuki ARAI, Koji Kamakura, Kenichiro Yashiro
Thirukkumaran Sivahumaran, Khee Pang "We propose using turbo code for the embedded
The aim of this paper is to analyze and compare the transmission (ET) scheme in optical code division multiple
performance of trellis coded Space Time Transmit Diversity access (CDMA). Although the scheme uses pulse position
(STTD) scheme and trellis coded Time Switched Transmit modulation (PPM), to resolve the problem of the
Diversity (TSTD) scheme for CDMA system. In the first part vulnerability of the embedded PPM signaling to multiple
of the paper we will derive an upper bound for the Frame access interference (MAI), we obtain soft information of the
Error Rate (FER) of both the schemes. We then use the received signal, which is used for turbo decoding. We
expressions for the upper bound to present the design of evaluate the bit error rate (BER) of the proposed scheme,
trellis codes that are optimum in terms of FER. In the compared the conventional PPM-OCDMA with turbo coding
second part of the paper we provide numerical results that and the ET scheme with BCH coding. Simulation results
validate the design. We conclude with a comparison of both show that the conventional turbo-coded PPM-OCDMA and
schemes under different channel conditions and multi-user than the BCH-coded ET scheme."
interference.
1C-09.4 1568965272
1C-09.2 1568965190
Impact of Propagation Parameters on
Blind Adaptive Multiuser Detection for Chaos Neighbouring Cells Interference in Power
CDMA Communication Controlled CDMA Systems
Lau Yuu Seng, Jusak Jusak, Zahir Hussain Samad Kolahi
"Chaos-based CDMA communication has opened up a new In this paper the effect of propagation parameters on
category for spread spectrum communication, where the neighbouring cells interference is investigated using a
spreading code that is used to scramble data in the CDMA simulation and modelling technique. The results of this
systems can be generated using a single mathematical investigation indicate the extent that other-cells interference
relationship of a chaotic generator instead of using the increases with increasing shadowing parameter but
pseudo-noise (PN) generator. Recent studies showed that decreases with increasing distance exponent. It is
this sequence can enhance the security and increase the concluded that the distance exponent has the dominant
capacity of the system at the physical link due to its effect on other-cells interferences. It is established that it is
bifurcation behavior. In this paper, we investigate the sufficient to consider only three tiers of neighbouring cells
performance of multiuser detection schemes in chaos when calculating the total interference as the addition of
based and PN code-division multiple access (CDMA) further tiers had an insignificant effect on the results. Other-
systems. Two well-known schemes are applied and cells interference factors are found to be independent of the
compared: adaptive filtering using the constant modulus traffic load when users are uniformly distributed.
algorithm (CMA) and matched filtering (using a correlator).
Simulation results showed that, for both schemes, the
chaos-based CDMA outperforms the PN-based CDMA in
terms of bit error rate (BER) and mean-squared error
(MSE)."

Tencon 2005 Page 50

 1C-10
Session Title: Power System Planning and Control 2
Chairperson: Ramesh Rayudu
Day: Tuesday
Date: 22nd November 2005
Time: 2:00PM – 3:40PM
Room: M10

1C-10.1 1568965120 1C-10.4 1568965264

A Novel Scheme of Load forecasting pertaining A Real-Time Monitor Framework for Static
to long term planning of a distribution system Voltage Stability of Power System
Ashvini Chaturvedi, Mandalaparty Murthy, Rakesh Jinli Zhao, Yu Yixin, Huiling Li, Jia Hongjie
Ranjan, Krishna Prasad "This paper proposes a framework for real-time static
"In this paper, a novel approach is proposed to address the voltage stability analysis and monitoring of bulk power
load-forecasting and expansion-planning process focused systems. This framework involves Continuous Power Flow
on long-range aspects. Depending upon land-use, five (CPF) studies, weak nodes and weak regions identification,
different user classes are considered. The existence of critical cut-set (transmission bottlenecks) selection and
these classes and their associated boundaries are modeled visualization of voltage stability regions. It aims to monitor
with trapezoidal distribution. A piece-wise linear model is voltage stability of power systems by judging whether the
used to model the growth ramp segment of S-shaped current operation point lies inside a considered static
growth characteristic. Finally two indices are proposed, one voltage stability region (SVSR) and calculating the voltage
to estimate the new load demand, and other index to stability margin in cut-set power space. The software
estimate the growth. The results obtained match the typical package based on this framework has been being
growth trend of different user classes for a practical developed in recent years and has basically taken shape till
distribution system." now. We will introduce the main framework and the
functions of this system."
1C-10.2 1568965370
1C-10.5 1568964796
THE enactment of optimal target value in power
plant operation based on language association Adaptive Setting of Digital Relay for
rule Transmission Line Protection
Xi-ning Yu, Cheng-lin Niu, Jian-qiang Li Jong-Man Cho, Chang-Ho Jung, Jin-O Kim
"This paper proposed language association rule and "Distance relay is the most widely used in transmission line
applied it to mining optimal target value in power plant protection because it is useful not only as main protection
operation. In language association rule, quantitative but also as back-up protection. However, the protection
attributes are partitioned into several fuzzy sets by FCM range of the distance relay is always fixed in the
algorithm, then the algorithm for mining Boolean association unchangeable operating range while the power system
rules is improved to discover frequent attributes. Last, varies, and therefore, the distance relay is the most highly
association rule with least confidence are generated to influenced by power system changes. In this regard, this
decide the optimal target value of operation parameters. paper describes an approach to minimize the mal-operation
Experiments and theory calculating show that this algorithm of the distance relay due to the power system changes
is effective and practical." through changing the protection range of the distance relay
into an optimal condition in response to load variation and
1C-10.3 1568965141 power system conditions. Also mal-function of the distance
relay in case of high resistance ground faults can be
An Analytical Method for the Reliability minimized through modeling the protection range into
Evaluation of Wind Energy Systems quadratic function."
Arash Ehsani, Mahmud Fotuhi-Firuzabad, Ali
Abbaspour, Ali Mohammad Ranjbar
"Wind generation is one of the most successful sources of
renewable energy for the production of electrical energy.
The technical characteristics of wind generation make
existing conventional generation models not directly
applicable. This paper presents an analytical method for the
reliability assessment of the flat-rated wind turbine
generation systems. An example of flat rating is that of the
MOD-2, a second-generation class of wind turbines. The
power-velocity characteristic of the flat-rated wind turbines
is employed in this paper to model the operating behavior of
the installed wind turbine generators. For wind-power
potential estimation, the Weibull distribution model is used.
The performance of the developed method is demonstrated
with computational results."

Tencon 2005 Page 51

 1C-11
Session Title: Miscellaneous Modelling
Chairperson: Enn Vinnal
Day: Tuesday
Date: 22nd November 2005
Time: 2:00PM – 3:40PM
Room: M7

1C-11.1 1568965689 1C-11.3 1568966664

Basket Bloom Filters for Membership Queries Characterisation of Combat Identification
Kun Xie, Min Yinghua, Da-Fang Zhang, Gao-Gang Technologies
Xie, Jigang Wen Cameron Boyd, Shaun Wilson, David Krause, Jim
"A Bloom Filter is a space-efficient data structure allowing Godfrey
membership queries over sets with allowable errors. It is "Network centric warfare combines sensors,
widely used in databases, networks, and distributed communications, and computation systems with tactics,
systems. This paper presents a novel Bloom Filter, called techniques and procedures to provide battlespace
Basket Bloom Filter (BBF). The BBF deals with different awareness. An element of this awareness is the ability to
elements in a data set depending on their query invalidation identify friendly combat entities, known as Combat
cost, by clustering elements into different baskets. The total Identification (CID). This paper describes current
query invalidation cost function is defined. In order to technologies used in combat identification and defines
minimize the total query invalidation cost, the genetic technology groups that will play a role in both current and
algorithm is employed to find the optimal number of hash future CID implementation."
functions for every basket. Simulation results show that, the
BBF has 40% lower total query invalidation cost than the
standard Bloom Filters under the same executing time."

1C-11.2 1568965782
An LMI Approach to Exponential Stability
Analysis of Neural Networks with Time-Varying
Delay
Wu-Hua Chen, Wei Xing Zheng
"This paper focuses on the problem of delaydependent
stability analysis of neural networks with variable delay.
Two types of variable delay are considered: one is
differentiable and has bounded derivative; the other one is
continuous and may vary very fast. By introducing a new
type of Lyapunov-Krasovskii functional, new delay-
dependent sufficient conditions for exponential stability of
delayed neural networks are derived in terms of linear
matrix inequalities. We also obtain delay-independent
stability criteria. These criteria can be tested numerically
and very efficiently using interior point algorithms. Two
examples are presented which show our results are less
conservative than the existing stability criteria."

Tencon 2005 Page 52

 1C-12
Session Title: Array and Space-Time Signal Processing 2
Chairperson: Lei Shang
Day: Tuesday
Date: 22nd November 2005
Time: 2:00PM – 3:40PM
Room: M9

1C-12.1 1568965133 1C-12.3 1568965313

Alamouti Coded OFDM in Rayleigh Fast Fading Accurate Performance Analysis of Optimum
Channels - Receiver Performance Analysis Space Diversity Combining and Equalization in
Himal Suraweera, Jean Armstrong Digital Point-to-point Radio
"In this paper, the receiver performance of Alamouti coded Sheng-Chou Lin, Hung-Hsing Lin, Hung-Wen Fang
orthogonal frequency division multiplexing (OFDM) systems "This paper is devoted to the fast and accurate performance
is analyzed. When the channel is not constant during the assessment of a digital point-to-point radio which uses high-
period of Alamouti codeword transmission, the conventional order modulaton quadrature amplitude (QAM) with optimum
linear maximum likelihood (ML) receiver suffers from space diversity combining and decision-feedback
performance degradation. We use three different receiver equalization. A comparison with linear equalization is also
combining methods and study the error performance. The provided. In our Monte Carlo simulation, the accurate error
sensitivity of different receivers is investigated by varying rate for each time-varying channel is obtained by using the
the correlation coefficient. Our results can be easily Gaussian quadrature rules (GQR) based on computation of
extended to be applicable to other block coded OFDM moments instead of the upper bound approximation. The
schemes with different antenna configurations. The use of GQR can reduce the calculating amount and
performance loss in linear ML receiver can be partially accelerate the speed of simulation significantly. The GQR
improved by employing a decision-feedback strategy. In the method can be applied to non-Gaussian noise or
decision feedback steps preliminary decisions are used to interference and always can assure accurate results and
subtract the intercodeword coupling. However this very satisfactory performance when the series expansion
increases the computational complexity of the receiver." method fails."

1C-12.2 1568965276 1C-12.4 1568965448

On the Design of Space-Frequency Codes from A Space-Time Block-Coded PCC-OFDM Scheme
Super-Orthogonal Space-Time Trellis Codes via for Flat Fading Wireless Channels
Mapping
Fawaz AL-Qahtani, Zahir Hussain
Kampol Woradit, Siwaruk Siwamogsatham, "This paper presents a transmission scheme joining
Lunchakorn Wuttisittikulkij Polynomial Cancellation Coding OFDM (PCC-OFDM) with
"This paper describes how to construct a class of Space- Space-Time Block-Coding (STBC). PCC is a coding method
frequency (SF) codes by mapping the Super-orthogonal for OFDM which maps data onto adjacent subacarriers,
space-time trellis (SOSTT) codes across OFDM used here for reducing the sensitivity of OFDM to carrier
subcarriers. Two mapping methods are employed: frequency offset and Doppler spread. On the other hand,
conventional repetition mapping and the proposed mapping STBC is used to improve the error performance and
methods with the design criteria of achieving full diversity increase the data rate of the transmission system.
including spatial and frequency at the maximum possible Simulation results have shown that the joint application of
transmission rate. The proposed mapping method appears PCC with OFDM can significantly improve the overall
to outperform the conventional mapping counterpart with system performance."
respect to coding gain when constructing the SF codes
mapped from SOSTT codes. Simulation results show the 1C-12.5 1568964881
performance gain of 0.5-3 dB under frequency-selective
fading channels. In addition, the constructed SF codes
Output-coding and SVM for multiclass microarray
mapped with both mapping methods are compared to the classification
existing commonly known SF codes, i.e. the Space-time Li Shen, Eng Chong Tan
trellis (STT) codes mapped with conventional mapping
method. Apparently the SOSTT codes are superior to the "Multiclass cancer classification based on microarray data
STT codes in single-carrier narrowband wireless channel, has been studied in this paper. A generalized output coding
the mapped SOSTT codes do not offer the same advantage scheme combined with support vector machines as binary
over the mapped STT codes under the severely-frequency- classifers is used. Different coding strategies, decoding
selective fading channel when the codes having large functions and feature selection methods are combined and
number of states. Nevertheless, some performance validated on two cancer datasets: GCM and ALL. By using
improvement is observed for codes with small number of the random coding strategy and recursive feature
states." elimination, the testing accuracy that we have achieved is
as high as 80.4% on the GCM data which has 14 classes.
Comparing with the other classification methods, our
method has shown its superiority in classificatory
performance."

Tencon 2005 Page 53

 1D-01
Session Title: Security Issues 1
Chairperson: Phillip Branch
Day: Tuesday
Date: 22nd November 2005
Time: 4:10PM – 5:50PM
Room: P2

1D-01.1 1568962776 1D-01.4 1568964174

Trust Models and Security Considerations in A Traitor Tracing Scheme Based on the RSA
Multicast Listener Discovery Protocol version 2 Scheme
(MLDv2) Yang Bo
Gopakumar Kurup, Greg Daley, Ahmet Sekercioglu "Traitor tracing schemes constitute a very useful tool
"Multicast Listener Discovery (MLD) protocols are employed against piracy in the context of digital content broadcast. In
by IPv6 routers and hosts for group management. Routers such multi-recipient encryption schemes, the data-suppliers
use MLD to discover the presence of multicast listeners - can reveal the identities of the subscribers that were
nodes that wish to receive multicast packets- on their implicated in the construction of a pirate-device that illegally
directly attached links, and to discover which multicast receives the digital content. In this paper, a traitor tracing
addresses are of interest to those neighbouring nodes. scheme based on the RSA scheme is proposed. The
Existing MLDv2 protocol specification discusses the effects scheme does not rely on any trusted third party, and is
of on-link forgery of MLD packets but does not provide any collusion resistant, its security is the same as that the RSA
protection from on-link attacks. By taking advantage of or scheme."
abusing MLD messages, bogus devices can cause
incorrect multicast records and disruption to multicast or 1D-01.5 1568964624
unicast packet delivery. In this paper we identify and
analyse the various trust models for MLD protocols, their
An Unconditionally Secure Multi-round
workings and their interaction with link-layer and multicast Revocation Scheme Using Secret Sharing
proxy devices. We also provide security and threat analysis Ming Yang
for each model."
"The revocation schemes are usually used for broadcast
1D-01.2 1568963666 encryption and multicast key management. We present a
Shamirs polynomial secret sharing based revocation
AutoMAV - Micro Aerial Vehicles for Airport scheme that enable many revocation rounds and has
Surveillance revocation capability in each round. Our scheme has some
notable properties as compared with the other precious
Stefan Winkler schemes: (1) Our multi-round revocation scheme is
"This paper gives an overview about the current status of unconditionally secure other than computationally secure.
the Micro Aerial Vehicle (MAV) project AutoMAV (2004- (2) Our scheme promotes the revocation capability from
2006). This project is being conducted by the leading users to users in every round, and is secure against any
German MAV research institutions and led by the Institute coalition of all the revoked users. (3) Each user only
of Aerospace Systems of the TU Braunschweig. The goal of restores the constant personal keys during the whole
the project is the development of a very agile, fully session, and the overhead of computation and
autonomous fixed-wing outdoor MAV for observing airports communication is only dependant on the degree of the
and other critical facilities in order to decrease the danger of secret sharing polynomial regardless of the size of the
terroristic acts." group."

1D-01.3 1568963667
Multiple Sensor Fusion for Autonomous Mini and
Micro Aerial Vehicle Navigation
Stefan Winkler
"Micro Aerial Vehicle (MAV) describes a category of aircraft
with dimensions roughly comparable to small birds. As the
smallest, powered aircraft, MAVs can carry various sensors
as payload to support such civil and military missions as
traffic monitoring, weather observation, and enemy
surveillance during military conflicts. In order to carry out
more sophisticated missions and to improve the flight
performance of the aircraft, a reliable navigation solution,
especially of the attitude, is of significant importance. For
this, an Inertial Measurement Unit (IMU) based on micro-
electromechanical systems (MEMS) were tightly-coupled
(usage of GPS raw data) with GPS. Subject of the paper is
the influence of the distance between IMU and GPS
antenna, the position of the GPS antenna relative to the
IMU and the number of GPS antennas used on the
navigation accuracy (position, velocity, attitude). The
simulation results will be verified in test flights."

Tencon 2005 Page 54

 1D-02
Session Title: Internet Applications 2
Chairperson: Grenville Armitage
Day: Tuesday
Date: 22nd November 2005
Time: 4:10PM – 5:50PM
Room: M1

1D-02.1 1568965056 1D-02.4 1568965232

A Late Join Approach for Distributed Dynamic- Rendering Models for Immersive Voice
locking in Real-time Collaborative Editing Communications within Distributed Virtual
Systems Environment
Peidi Ji, Liang Cai, Chun Chen, Bo Jiang Ying Que, Paul Boustead, Farzad Safaei
"This paper proposes a novel late join approach for "This paper compares three possible rendering models for
Dynamic Locking Mechanism (DLM) in Internet-based real- the provision of Immersive Voice Communications (IVCs) in
time collaborative editing systems. The main idea is that: in Distributed Virtual Environments (DVEs) such as
the process of late join, four cases are classified according multiplayer online games. The common aim of these three
to the situations of whether the user has enabled the DLM rendering models is to create a personalised auditory scene
or not and whether the user is in the session or not. And for each listening avatar, consisting of a mix of the
different transmission steps are adopted for each case in surrounding avatars voices, positioned according to their
order to achieve the goal that the dynamic locking data can positions in the virtual world. The first two rendering models
be synchronized. This approach can maintain the are based on amplitude panning localisation and HRTF-
consistency of locking data in the distributed collaborative based binaural localisation respectively. The computation
environment more efficiently. And the process of late join for cost of the latter is deemed too large to meet the identified
multiple users and the error control method in the network processing power constraints. A computation reuse scheme
transmission are also presented in this paper." was introduced in the third rendering model which, as
shown in our simulation results, reduces significantly the
1D-02.2 1568965077 computational cost of providing IVC using HRTF-based
binaural localisation."
Support Tool Using Positions of Contents for
Voice/Gaze-controlled Browsers 1D-02.5 1568965258
Masahide YUASA, Minoru Ohyama An Enhanced Framing Strategy for Jitter
"In this paper, we propose a support tool termed Calamari Management
to improve voice/gaze-controlled browsers by using
information about the position of contents. It can quickly
Jhoanna Rhodette Pedrasa, Cedric Angelo Festin
determine the positions of links, images, and tables in a "The Internet Protocol's best-effort service is ill-suited for
web page. Positions in a browser window are important for delivering real-time, interactive applications such as voice.
voice/gaze-controlled browsers to deal with user commands Additional management is needed to maintain quality at
associated with directions, for example, select ``upper one"" satisfactory levels throughout the duration of the call. A new
and ``left figure."" By using positional information, framework at network management, the Value-Based Utility
developers can effectively improve these browsers to use model, uses the knowledge of user satisfaction levels in
commands along with directions. The proposed tool is order to allocate resources more efficiently. Our research
implemented in a proxy server; hence, it does not depend applies the Value-Based Utility model to jitter management
on browsers and operating systems. Although it rewrites an and demonstrate that performance improvements can be
HTML document, it does not change the web page design. achieved by enhancing on the frame-based scheduling
We describe two applications using the tool and algorithm Stop-and-Go. Our results show that Stop-and-Go
demonstrate that it is useful for developers to enhance the with Utility can achieve a tighter jitter bound of T, where T is
voice/gaze-controlled browsers." the frame length, which is half of the original bound of 2T
offered by Stop-and-Go alone."
1D-02.3 1568965222
An Approach for OWL-Based Web Services
Components and Composition Framework
Jianjing Shen, Huijie Zhao, Wang Fengsi
"By using component and framework technology, in this
paper, we present our current work on the concepts and
technologies of designing, constructing, and composing
OWL-based Web Services components. A framework for
composing Web Services components is proposed, and
workflow management of a Semantic Web Services based
on the proposed framework is provided."

Tencon 2005 Page 55

 1D-03
Session Title: Robotics
Chairperson: Raymond Jarvis
Day: Tuesday
Date: 22nd November 2005
Time: 4:10PM – 5:50PM
Room: P1

1D-03.1 1568964833 1D-03.3 1568965181

Robot Path Planning in High Risk Fire Front Accommodating Uncertainty in Covert and Overt
Environments Robot Path Planning
Raymond Jarvis, Mohamed Marzouqi Mohamed Marzouqi, Raymond Jarvis
“Most path planning approaches for mobile robot navigation "In this paper, the notion of probabilities has been
have concentrated on discovering obstacle-free paths which introduced to plan a visibility-based path to adapt with real
minimise distance, time or energy. Here we show how world uncertainties. An initially known or suspected entities
variations adapted for Covert Robotics [1,2,3], where locations, whose the robot needs to either minimize (covert)
visibility from various (or all) points in the environment is or maximize (overt) its exposure to them, can be
built into the cost structures, can be further adapted to inaccurately estimated or they may unpredictably moving.
provide robot navigation paths which minimise risk of fire Promising results have been shown when testing the
heat exposure in high risk fire front environments. Both planning technique on a simulated environment. A number
escape routes and fire extinguishing forays can be planned of test cases are presented for both covert and overt
using this methodology. A wide range of simulation results applications."
are presented to illustrate the power and scope of this
approach. The work reported is related to an Australian 1D-03.4 1568965311
Research Council Linkage Grant project entitled Robotic
Vehicles for Fire Fighting and Emergency Services Bluetooth Based Proximity Sensing for Reactive
Support." Mobile Robots
Robert Shepherd, Sa'ad Mansoor
1D-03.2 1568965115
"The Bluetooth radio system is an ideal communications
Recovery from Segmentation Failures Using tool for lightweight autonomous mobile robots but little
Photometric Invariance in an Interactive Object research interest has been targeted at this application.
Recognition System Along with facilitating ad hoc communications, other
features of Bluetooth may be exploited for use in the mobile
Md. Altab Hossain, Rahmadi Kurnia, Yoshinori Kuno, robotics domain. We consider the likely scenario where a
Akio Nakamura mobile robot becomes inoperably stuck and unable to free
"We are developing a helper robot that carries out tasks itself. The robot should be able to warn other robots about
ordered by the user through speech. The robot needs a its situation in order to prevent further losses from the same
vision system to recognize the objects appearing in the potential failure. In this paper we demonstrate how the
orders. It is, however, difficult to realize vision systems that discovery phase of Bluetooth connection setup can provide
can work in various conditions. Thus, we have proposed to simple proximity data without requiring an explicit
use the human user's assistance through speech. When the connection to be established. Our lightweight proximity
vision system cannot achieve a task, the robot makes a sensing mechanism, tightly coupled to a behaviour based
speech to the user so that the natural response by the user control architecture provides a cheap but effective method
can give helpful information for its vision system. Our for kin recognition. Using the latency of the discovery
previous system assumes that it can segment images process, we show various proximities can be attained whilst
without failure. However, if there are occluded objects minimising power usage."
and/or objects composed of multicolor parts, segmentation
failures cannot be avoided. This paper presents an 1D-03.5 1568965401
extended system that tries to recover from segmentation Designing of Human Computer Interactive
failures using photometric invariance. If the system is not
sure about segmentation results, the system asks the user
Platform for Robotic Applications
by appropriate expressions depending on the invariant Manas Bhuyan, Debashis Ghosh, Prabin, Kumar
values. Experimental results show the usefulness of the Bora
system."
"The use of human hand as a natural interface for human-
computer interaction (HCI) serves as the motivation for
research in hand gesture recognition. Gestures provide a
rich and intuitive form of interaction for controlling robots.
Vision-based hand gesture recognition involves visual
analysis of hand shape, position and/or movement. This
paper describes a system which enables a robot to
recognize and respond to hand gestures from a human
operator. The interface uses some predefined hand
gestures, each of which represents an individual control
command for the motion control of the remote robot.
Experimental results demonstrate that our proposed HCI
platform can be used reliably in Robotic applications."

Tencon 2005 Page 56

 1D-04
Session Title: Microwave Circuits
Chairperson: Ernest Farbin
Day: Tuesday
Date: 22nd November 2005
Time: 4:10PM – 5:50PM
Room: P3

1D-04.1 1568963318 materials fundamental behaviour. On the other hand, it has

been interesting to note that the temperature coefficient of
Low cost intergidital BST varactors for tunable permittivity of a material is a constant irrespective of the
microwave applications measurement technique. The studies on temperature
Ernest Fardin, Kamran Ghorbani, Anthony Holland dependent permittivity will reveal the knack of a material to
operate as a frequency stable circuit or device."
"In this paper, a simple and inexpensive technique for the
fabrication of barium strontium titanate (BST) varactors is 1D-04.4 1568965005
presented. BST films were deposited using RF magnetron
sputtering on platinum coated silicon substrates. The Planar Ba(MgxTay)O3 Material for Emerging
advantages of silicon include low cost and ease of Microwave Technologies
integration with existing fabrication processes. The
microstructural properties of the BST layer were examined
Mohan Jacob, Janina Mazierska
with scanning electron microscopy (SEM), x-ray diffraction Ba(MgxTay)O3 based dielectric materials are typically used
(XRD) and x-ray photoelectron spectroscopy (XPS) in many microwave applications due to its relatively high
techniques. Gold top layer electrodes were deposited using permittivity and low temperature coefficient of frequency.
DC sputtering and patterned with a conventional MURATA® manufactured a dielectric substrate based on
photolithographic process. The S-parameters of the Ba(MgxTay)O3; we have characterised this substrate at
varactor were measured from 40 MHz to 20 GHz on a microwave frequencies and at cryogenic temperatures
vector network analyser. From this data, the capacitance, using cryogenic Split Post Dielectric Resonator. The TE011
tunability and Q-factor of the device were extracted. The mode of this resonator without and with sample is identified
BST-based varactors show potential for integration into at frequencies of 10 GHz and 8.6 GHz respectively. The
electronically tunable microwave circuits." Transmission Mode Q factor Technique was used for data
processing to ensure high accuracy of the measurements.
1D-04.2 1568964675 The real part of the permittivity and the loss tangent varies
between 24.37 and 24.4 and 4.2x10-5 and 6x10-5
High Speed Over Ocean Radio Link to Great respectively in the temperature range 25 295 K. The
Barrier Reef variation in permittivity is less than 0.2%. The low
Christopher Palazzi, Graham Woods, Ian Atkinson, temperature coefficient of frequency and permittivity make
Ba(MgxTay)O3 substrate attractive to fabricate miniaturised
Stuart Kininmonth
planar antenna arrays and in other microwave applications
"This paper investigates the feasibility of using high speed where frequency stable substrates are needed.
microwave radio links to relay information from the Great
Barrier Reef to the Australian mainland. The investigated 1D-04.5 1568965076
solution is a low-elevation, microwave link operating at
10.5GHz. Radio signals in this band are known to often Dual Coupled Resonator Local Oscillator
become trapped in the evaporation duct just above the Cornelis Kikkert
ocean providing a suitable means to radiate signals well
beyond the optical horizon. The paper describes "This paper describes how two coupled resonators, thus
experimental and prediction results for a 78km link between forming a second order coupled network can be used to
Davies Reef and the Australian Institute of Marine Science. produce an oscillator with a lower phase noise than can be
The optimum frequency and antenna heights are obtained using a single resonator. The design technique is
determined and the expected variability in the received illustrated using a typical MMIC and Microstripline
signal level is determined. It is predicted that using resonators. Microwave Office computer simulation shows a
conventional radio equipment a received signal to noise 2.8 dB improvement in phase noise. The constructed
ratio of 40dB should be achievable." oscillator oscillated close to the design frequency and
produced a lower phase noise than a single resonator
1D-04.3 1568965004 oscillator using the active same devices."

"Are the parameters Q0×f0, temperature

coefficient of frequency and temperature
coefficient of permittivity fundamental material
constants?"
Mohan Jacob
"Microwave characterisation of dielectric materials is vital to
implement the material in communication devices and
circuits. The microwave characterisation parameters are the
dielectric constant, loss tangent, Q0×f0, temperature
coefficient of frequency and temperature coefficient of
permittivity. From several experiments on different dielectric
materials, it has been concluded that the concept of using
Q0×f0 and temperature coefficient of frequency, as a
material property was wrong since it is not reflecting on a

Tencon 2005 Page 57

 1D-05
Session Title: Speech Signal Processing and Coding
Chairperson: Tarashankar Rudra
Day: Tuesday
Date: 22nd November 2005
Time: 4:10PM – 5:50PM
Room: M2

1D-05.1 1568965023 1D-05.3 1568964207

Generation of Excitation Signal in Voice excited A variable low bit rate Speech coder based on
Linear Predictive Coding using Discrete Cosine MELP
Transform Ding Qi
Gaurav Khare, Prashant Shekhar, M. Kulkarni "This paper proposes a novel variable low bit rate speech
"Methods involved to generate the excitation parameters coding algorithm based on Mixed Excitation Linear
and the gain have been proposed that provide Predictive (MELP) coder. This coder use a superframe
improvements over plain Linear Predictive Coding (LPC) in structure and joint quantization methods to reduce the inter-
terms of better special reconstruction. DCT computation to frame redundancy of the parameters. In each superframe,
transmit the excitation signal energy in initial few the similarity of the linear prediction filter denoted by LSF
coefficients gives a better reconstruction of innovation parameters is imposed to reduce the bit rate by variable
signal at the receiver as compared to the primitive method rate speech coding. A new distance measurement for LSF
of pitch estimation. Gain computation in terms of the Root is proposed to describe the difference between two sets of
Mean Square (RMS) value of the voltage levels for each LSF parameter. And the coding rate is decided by the
frame reduces the complexity involved as compared to plain distance measures of different LSF parameters in every
LPC, where gain was calculated in terms of mean square superframe. Objective test results of synthesized speech
error." quality show that the proposed speech coder is capable of
maintaining acceptable speech quality at the rates 560 bps
1D-05.2 1568965347 to 800 bps."

VLSI Architecture Design for Concatenative 1D-05.4 1568962328

Speech Synthesizer
Cognitive Emotion in Speech Interactive Games
Jia-Ching Wang
Tarashankar Rudra
"This paper presents a VLSI architecture for Mandarin
speech synthesis. For the natural synthesized speech, "This paper falls into the category of Human Computer
subsyllable based synthesis units are recorded in advance. Interaction. Here the focus lies in injecting emotions into the
The synthesized speech is obtained by suitably non player characters (NPC) in reaction to the emotion of
concatenating the synthesis units. The TD-PSOLA (Time the player in computer games. The idea of injecting emotion
Domain Pitch Synchronous Overlap-and-Add) approach is into games has been a popular topic for discussion in the
used to perform the prosody modification. The proposed Games industry today. The paper reviews the emotional
VLSI architecture includes two parts: the TD-PSOLA side of games from both the game side and the human
module and the synthesized pitch period generator. In the player side, considering speech style and content, facial
TD-PSOLA module, we also present a fast CORDIC expressions and gestures. The emotional content is
architecture which is five times faster than the conventional encapsulated within an XML [1] and an associated finite
method." state automaton."

Tencon 2005 Page 58

 1D-06
Session Title: Power Transformers
Chairperson: Palani Anniyappan
Day: Tuesday
Date: 22nd November 2005
Time: 4:10PM – 5:50PM
Room: M3

1D-06.1 1568964085 1D-06.4 1568965345

Minimization of transmission loss with Minimum Cost Design of Small Low-Loss
consideration of transformer taps constraint Transformers
Somchai Biansoongnern, Songsak Chusanapiputt, Teeraphon Phophongviwat, Chaiwut Chat-uthai
Sukumvit Phoomvuthisarn "A procedure for minimum cost design of a small low-loss
"This paper describes the solution of minimization of low-frequency transformer is presented. The total losses of
transmission loss with consideration of transformer taps transformer can be defined to be the design constraint. The
constraint. The transformer taps are using as control procedure for improving the accuracy of calculated total
variables for transmission loss minimization and maintain losses is the basis of the proposed total losses calculation
voltage profile when the disturbance occurs in system. associated with the temperature rise evaluation. The
When transformer taps are only used for minimization of optimization technique used in this paper is based on the
transmission loss that has to control transformer taps close Genetic Algorithms. Five samples of 72 VA, 220/24 V, 50
to center. Proposed methodology is to control the stages of Hz transformers have been designed and tested in order to
transformer taps close to the center of transformer taps. illustrate how well the calculated total losses values using
The equation of transformer taps stages is included in the this proposed procedure matches that of the measured
objective function and in inequality constraints. An optimal values obtained from the test. Finally, the simulation results
power flow problem is solved by using interior point method. of the minimum cost design with losses constraint are
The problem constraints are the power flow equation, the demonstrated. This proposed design is suitable for the
bound of the power system and stages of transformer taps. design of economical transformer with total losses value
The performance of the method is illustrated in IEEE 14 bus options which is considerably flexible for a particular
IEEE 30 bus and IEEE 57 bus test system. Test results purpose"
show that the transmission loss can be made minimum and
transformer taps can be controlling close to the central tap." 1D-06.5 1568965405
An Adaptive High Resolution Measuring System
1D-06.2 1568965060
for Characterizing Transformers Subjected to
Research on Novel Power Transformer Broadband Excitation
Protection Based on Self-Correction Function
Palani Anniyappan, Jayashankar Venkataraman
Yan XU Jing MA, Zengping WANG "Several phenomena in transformers are best analyzed in
"When differential protective relay is applied to transformer the frequency domain when energized by broadband
protection, how to identify the magnetizing inrush current excitation. These include dielectric behavior to lightning
and short circuit current caused by internal fault is the key impulse, winding deformation during transport and structural
problem. Here, regarding the inrush current and short circuit failure during short circuit tests. We propose an adaptive
current as the random signal, the sampled data are architecture that characterizes all these events in an
analyzed by correlation function principle in the digital signal unifying manner. The developed instrument is based on the
processing £¨DSP£©and the self-correlation function of the PXI hardware with necessary software developed under
sampled data is calculated. Through the similar comparison LABVIEW environment. Experimental results using the
operation conducted by standard self-correlation function developed instrument are reported to demonstrate its
(SSCF) formed with sinusoidal current the short circuit versatility."
current and the inrush current can be distinguished by the
magnitude of similarity coefficient. The results in Electrical
Power Dynamic Laboratory (EPDL) verify the feasibility of
the proposed method."

1D-06.3 1568965062
A Novel Inductance Calculation Method in Power
Transformer Model Based on Magnetic Circuit
Yan XU, Zengping WANG, Qing LIU
"For the problems in the inductance calculation of
transformer simulation model, this paper presents a
calculation method with transformer magnetic circuit. This
method calculates the mutual inductance and leakage
inductance respectively. When calculating the leakage
inductance, the effect of vertical leakage flux and horizontal
leakage flux have been considered, which makes the
calculating result are more correct. In order to prove the
validity of this method, the data of simulation test and
dynamic analog test have been compared. The results have
proved that the method of magnetic circuit calculation is
accurate."

Tencon 2005 Page 59

 1D-07
Session Title: Electronic Devices 3
Chairperson: John Richard Hizon
Day: Tuesday
Date: 22nd November 2005
Time: 4:10PM – 5:50PM
Room: M4

1D-07.1 1568965331 1D-07.4 1568965420

Detection of Faulty Semiconductor Wafers using Development of an Analog Electronic Wattmeter
Dynamic Growing Self Organizing Map Mahfoozur Rehman, Shamshuddin Ahmad, Shamim
Azharul Karim, Georg Russ, Arthur Hsu, Aminul Ahmad Ansari, "Misbahur Rehman Saad
Islam, Saman Halgamuge, Alan Smith "The paper deals with the development of an analog
"Quickly solving product yield and quality problems in a Electronic wattmeter which is simple in design and can
complex manufacturing process is becoming increasingly measure power, from very low values to very high values,
more difficult. There are various types of failures and their with simple modifications, at different frequencies and with
causes have complex interrelationships. These problems good order of accuracy. It can also be used in the
include: multi-factor & nonlinear interactions; intermittent measurement of power of distorted signals and can be
problems; dynamically changing processes; multiple interfaced with out any loading error. Basically, it measures
products; and the increasing volumes of data. This huge I cos with good order of accuracy and effect of variation of
volume of data coupled with quicker time to market voltage is taken care off by introducing a voltage sensitive
expectations is making finding and resolving problems resistance. By introducing electronic circuits at different
quickly an overwhelming task. In this study, a data mining stages, phase angle errors due to imperfections can be
algorithm called dynamic growing self organizing map has compensated. Output voltage, at the final stage, will be
been applied to wafer manufacturing data to detect the proportional to average power of the system"
faulty products. A clustering quality measure was developed
to evaluate the performance of the algorithm in separating 1D-07.5 1568965427
good and bad products. Clustering quality 1 and 0 means
100% separation and no separation respectively. Results
Monolithic Spiral Inductors for a 0.25 um Digital
show that the algorithm was able to separate good and bad CMOS Process
product from the raw data Whereas this paper has focused Marc Rosales, John Richard Hizon, Louis Alarcon,
mainly on clustering good and bad products, the technique Delfin Jay IX Sabido
can be extended to model the failure causes of the lower
yielding products." "The continued scaling of the transistor in digital CMOS
processes has considerably improved its characteristic for
1D-07.3 1568965390 RF applications. This has triggered a lot of research on the
potential of using CMOS technology for wireless
Frequency and Flexibility Based Cell Phone applications. The main objective is the integration of both
Keypad Layout baseband circuits and RF blocks in CMOS resulting in
wireless devices that are cheap and battery friendly.
Mohammad Azad, Rezwana Sharmeen, Shabbir However, a major limitation in the integration of RF systems
Ahmad, Riyad Mahmud on CMOS is the lack of models that will accurately predict
"Currently available cell phone key pad is not scientific in the behavior of passive elements such as inductors. This
case of user friendliness and at times it is very difficult to presents a major limitation for the successful integration of
use. New approach to enhance cell phone keypad is made RF building blocks in CMOS. In this study, various
using the layout of pc based ""qwerty"" keyboard. Since it is alternatives in integrating inductors in CMOS are
not also a feasible solution this paper will suggest a key investigated. These inductors are characterized using an
board for cell phone and other cellular device based on the inductor model to effectively compare the merits of each
frequency of the alphabets in English language and also implementation and to identify relevant parasitics that limit
with the view of structure of human finger movements to the performance of these inductors."
provide a flexible and fast typing cell phone keypad layout"

Tencon 2005 Page 60

 1D-08
Session Title: Image Compression and Quality Assessment
Chairperson: Donald Bailey
Day: Tuesday
Date: 22nd November 2005
Time: 4:10PM – 5:50PM
Room: M6

1D-08.1 1568964903 1D-08.4 1568965111

Optimizing the FBI Fingerprint Compression An Improved Visual Pruning Algorithm for
Algorithm for BLACKFIN Processor Perceptually Lossless Medical Image Coding
Cong-Van Nguyen, David B. H. Tay, Guang Deng David Wu, Damian Tan, Hong Ren Wu
"This paper presents an effective method to convert the "An improved algorithm for perceptually lossless coding of
conventional floating-point algorithm of the FBI Fingerprint medical images is presented in this paper. Built on the
Compression into a fixed-point algorithm suitable for digital JPEG 2000 coding framework, the proposed coder
signal processors. Wavelet transform, the most combines an improved Visual Pruning algorithm with an
computationally expensive component of the technique, is advanced model of the Human vision system to identify and
optimized to reduce the computational burden significantly. remove visually insignificant/irrelevant information. Current
The conversion and optimization are tailored to fit the results have shown superior compression ratio gains over
advanced structure of the Blackfin processor. The novel that of its lossless counterparts without any loss in visual
algorithm also retains an acceptable quality of the fidelity."
reconstructed image. Index TermsFingerprint image
compression, wavelet scalar quantization compression, 1D-08.5 1568963065
Blackfin."
A New Automatic Gait Recognition method based
1D-08.2 1568965184 on the Perceptual Curve
Subjective Image Quality Assessment at the Yanmei Chai, Qing Wang, Rongchun Zhao,
Threshold Level Changzhu Wu
Chris White,Damian Tan, Jianfei Cai, David Wu, Chin “Recognizing people by their gait is an emerging biometrics.
In this paper we proposed a new automatic gait recognition
Soon Tan, Rodney Martin approach based on the perceptual curve, which can
"A subjective method for assessing image quality at the effectively preserve the temporal changes of the walker's
threshold level is presented. The assessment methodology silhouette shape consistency with human vision. Firstly, we
has a trichotomous (3-way) structure. It looks at some detect the walking person in each the monocular image
problems associated with subjective assessments, sequences. Then we use inner boundary tracking algorithm
particularly fatigue and scanning effects. In addition, it to extract the walker's binary silhouettes and compute the
addresses the elusive issue of the minimum sample size PSD (Perceptual Shape Descriptor) of its shapes. Next,
necessary to achieve statistical relevance in subjective accumulative PSD of each image sequence forms a
tests. This leads to the question of the reliability of subjects, perceptual curve as its gait signature. Finally, the STC
a problem which may be resolved with a robust assessment (Spatio-temporal Correlation) similarity measure and two
method." different simple classification methods (NN and KNN) are
used to recognize different subjects. Experimental results
1D-08.3 1568965573 on the UCSD database and CMU database demonstrate
the feasibility of the proposed approach."
Fast hybrid fractal-VQ color compression in
Wavelet domain
Madhuri Khambete, Madhuri Joshi
"Fractal image compression is a technique based on the
representation of an image by a contractive transform and
iterated function system. Resultant attractor of contractive
transform is close to the original image. Though pure fractal
compression cannot give compression ratio comparable to
the state of the art schemes, hybrid schemes incorporating
fractal compression and alternative techniques have
achieved good compression ratio. This paper presents
novel color image compression algorithm using fractal
compression and Vector Quantization, which utilizes
correlation between RGB components."

Tencon 2005 Page 61

 1D-09
Session Title: Estimation and Modelling 1
Chairperson: Zahir Hussain
Day: Tuesday
Date: 22nd November 2005
Time: 4:10PM – 5:50PM
Room: M8

1D-09.1 1568964895 1D-09.4 1568965279

Frequency Estimation of Mono- and Multi- Human Identification System Based ECG Signal
component FM Signals Using the T – Sukkharak Seachia, Jeerasuda Koseeyaporn,
Distributions Paramote Wardkein
Zahir Hussain "A new human identification system based
"In a recent work we have proposed a subclass of Cohen's electrocardiogram (ECG) signal is introduced in this work.
Class of quadratic time-frequency distributions (TFD's), the The human heart is considered to be a unique system of
T-class of distributions with time-only Doppler-lag kernels to each person. ECG signal therefore represents as an
provide high-resolution and considerable cross-terms impulse response of the system. The frequency response of
reduction for FM signals. In this work we investigate the the system (Fourier transform of the ECG signal) is
instantaneous frequency (IF) properties of two members of employed to be a tool for feature extraction. In addition, the
this class: the hyperbolic and the exponential T-distributions ECG signals employed in this paper, which may be derived
in the presence of Gaussian noise. Both mono- and multi- from different heart rates from different subjects at the
component FM signals will be considered, with various recording time, are normalized to a standard heart rate.
modulation coefficients. A comparison with two well-known Furthermore, not only the whole sequence of 1 period EC
TFD's, Wigner-Ville and Choi-Williams distributions, is signal, containing P, QRS, and T waves, is processed but
presented for performance evaluation." also its three subsequences, each respectively representing
P, QRS, and T waves, is examined. The results of using
1D-09.2 1568965132 neural network have demonstrated that subsequences
technique is superior to whole period of ECG signal
An Approach to Resolve Integer Ambiguity of method."
GPS Carrier Phase Difference for Spacecraft
Attitude Determination 1D-09.5 1568965284
Somphop Purivigraipong Fast and Accurate Positioning Technique Using
"This paper presents a new approach to resolve integer Ultrasonic Phase Accordance Method
ambiguity and achieve instantaneous spacecraft attitude Hiromichi Hashizume, Ayumu Kaneko, Yusuke
from phase information of global positioning system signals, Sugano, Koji Yatani, Masanori Sugimoto
without requiring independent attitude information from
traditional attitude sensors. A newly approached algorithm "In this paper we propose a new technology for measuring
based on Gram-Schmidt Orthonormalisation is proposed for precise distance/angle toward the terget using unexpensive
medium length baseline observations. Using an attitude ultrasonic transmitter/receiver transducers. The system
algorithm from vector observations, an instantaneous measures distance by the transmission delay time of an
estimated attitude is obtained from only four phase ultrasonic packet, and the key idea is that the packet is
measurements collected from only two baseline marked with a special point, epoch using two carrier
observations. The implemented algorithms are tested with a frequencies, and at the receiver end is analized and
flight data collected onboard a mini-satellite orbiting in low retrieved the epoch within a few microseconds' error range.
Earth orbit. The results showed that a coarse attitude We name this new technology as phase accordance
solution is achieved from phase information of stand-alone method, and demonstrate that the distances to objects of a
global positioning system signals." few meters range are measured within a millimeter. This is
a significant improvement in distance measurements using
1D-09.3 1568965139 ultrasonic sensors. The applications of the same technology
to angle measurements and data communication are also
Spacecraft Orbit Determination from Code discussed."
Information
Somphop Purivigraipong
"This paper presents the orbit determination of low-Earth
orbit (LEO) spacecraft from code information of global
position system (GPS) signals. An extended Kalman filter
was implemented to estimate the position and velocity of
the spacecraft. The effect of second harmonic, J2, of the
Earths oblateness was included in the system model. The
estimated results were evaluated by the reference solutions
emulated from the developed GPS data-generator. The
conventional least squares method was exploited to provide
the initial guess"

Tencon 2005 Page 62

 1D-10
Session Title: Informatics and AI
Chairperson: Horace King
Day: Tuesday
Date: 22nd November 2005
Time: 4:10PM – 5:50PM
Room: M10

1D-10.1 1568964087 receiver and also mutual trust (client verifies the identity of
the server and server verifies identity of the client)."
Reconfigurable Computing: Peripheral Power
and Area Optimisation Techniques 1D-10.4 1568964614
YiShian Lee, Tim Oliver, Douglas Maskell Intelligent Automated Health Systems for
"This paper examines techniques to achieve power, area Compliance Monitoring
and configuration storage reductions in the peripherals of a
reconfigurable computing system. The two techniques
Olaf Diegel
chosen are the propagation of constants to the peripheral "Intelligent health systems are good examples of the
core and the modification of the Finite State Machine (FSM) integration between users and technology. Elderly users
to reduce switching activity. The constant propagation have very specific requirements for health and safety. They
technique is based on the observation that configuration of constitute a special user group as they are strongly habit
peripheral devices is generally set only once during the focused and are thus good subjects for intelligent systems.
lifetime of the device. Since the peripheral is running only a Smart house computer management systems need to have
single function, the control registers now become intelligent systems in order to make users lifestyles safe
redundant, as there is only one input for each of the and healthy while, at the same time, giving them
registers. Removing this eliminates much of the redundant independence of living. This paper describes how such a
control registers and routines. FSM modification is based on system was implemented whereby the computer
the observation that signal transitions consume power. To incorporates an intelligent system which initially learns from
reduce the power consumption, the state representations in its users habits and then starts to make intelligent decisions
the FSM are replaced by Gray Code. Gray code only based on these patterns. This helps users better manage
changes one bit at a time, hence, power consumption can their health and medication compliance in a way that is
be reduced. Results show a reduction in area and seamless and transparent."
configuration storage of up to 20% and a power reduction of
up to 7%." 1D-10.5 1568965522

1D-10.2 1568964912 An Artificial Neural-Network-Based Approach to

Software Reliability Assessment
An Efficient Temporal Formula Specification
Method for Asynchronous Concurrent Systems Su Yu-Shen, Chin-Yu Huang, Yi-Shin Chen
"In this paper, we propose a neural-network-based
Chikatoshi Yamada approach to software reliability assessment. We first derive
"Design verification has played an important role in the mathematic expressions that can be applied to neural
design of large scale and complex systems. In this article, networks from traditional software reliability models. Next,
we focus on specification process of model checking. we show how to apply neural network to software reliability
Behaviors of modeled systems are in general specified by modeling by designing different elements of neural
temporal formulas of computation tree logic, and users must networks. Our proposed neural-network-based model can
know well about temporal specification because the combine various existing models to build a more common
specification might be complex. We propose a method by one. Finally, we compare the performances of our proposed
which temporal formulas are obtained inductively, and model with other models with three aspects: goodness-of-
amounts of memory, OBDD nodes, and execution time are fit, short-term prediction, and long-term prediction. Based
reduced. We will show verification results using the on the experimental results, our model significantly
proposed a temporal formula specification method." outperforms the traditional models."

1D-10.3 1568965072
Dynamic User Credential Management in Grid
Environment
Debasish Jana, Amritava Chaudhuri, Datta Abhijit,
Bijan Bihari Bhaumik
"In grid computing, heterogeneous computing stations
spread out in the world can connect to the grid and offer
services or request services in a loosely coupled
environment with services-on-demand style. Public Key
Infrastructure (PKI) is the most widely adopted security
infrastructure used in Grid environments. We have
developed a scheme for dynamic token generation in a grid
environment ensuring more security and less hack
proneness because of the dynamic changing of the token
which is used in all transactions. The dynamic token thus
generated forms part of the private key and the user id of
the client provides the public key in terms of PKI. The
dynamic token helps in establishing proof of origin to the

Tencon 2005 Page 63

 1D-11
Session Title: Conversion Issues
Chairperson: Maria Theresa Gusad
Day: Tuesday
Date: 22nd November 2005
Time: 4:10PM – 5:50PM
Room: M7

1D-11.1 1568963842 1D-11.3 1568965103

Design Methodology for CMOS Low-Noise Modeling and design considerations for zvs
Amplifiers Using Power Matching Techniques asymmetrical half-bridge converter
Maria Theresa Gusad, Louis Alarcon Guan-Chyun Hsieh, Chi-Yeh Kuo
"In this paper, a methodology in designing CMOS Low- "Practical point of view in modeling and design
Noise Amplifiers (LNAs) is proposed. Three power- considerations for a zero-voltage-switched (ZVS)
matching techniques are considered in the design of the asymmetrical half-bridge (ASY-HB) converter is proposed.
LNA. These are: (1) matching for maximum available gain, Six dynamic conversion states for describing the converter
(2) matching for a constant gain, and (3) matching for characteristics are explored and modeled, including two
stability. Using a 0.25 mm CMOS process, several LNA power-delivery states, two resonant states, and two energy-
circuits employing the common-source topology with transition states etc. Design considerations are examined
cascode configuration are designed, implemented, thoroughly. A design of 300W ZVS ASY-HB converter is
fabricated, and tested. The performance of LNA circuits examined and a high efficiency up to 92% at full load is
designed using the three different techniques are achieved. All performances verified by experiment are quite
characterized. Simulation and actual measurement results well and close to the theoretical predictions."
are also compared and analyzed to determine the capability
of the simulator to predict the LNAs overall performance at 1D-11.4 1568963323
radio frequencies."
Digital Control of High DC Voltage Converter
1D-11.2 1568965057 Based on Cockcroft Walton Voltage Multiplier
Circuit
A Mixed-Split Scheme for 2-D DPCM Based LSF
Quantization Saad Mekhilef
Saikat Chatterjee, Thippur Venkat Sreenivas A new method of controlling high DC voltage based on
Cockcroft Walton Voltage Multiplier circuit by using digital
"In this paper a mixed-split scheme is proposed in the controller is presented. The digital controller is developed
context of 2-D DPCM based LSF quantization scheme using Complex Programmable Logic Devices (CPLD). The
employing split vector product VQ mechanism. proposed system utilizes a single-phase AC as an input
Experimental evaluation shows that the new scheme is supply. The power switching devices in the controlled
successfully being able to show better distortion bridge are controlled by the multiple-pulse Pulse Width
performance than existing safety-net scheme for noisy Modulation (PWM) switching technique so as to minimize
channel even at considerably lower search complexity, by the low order harmonic present on the AC side of the
efficiently exploiting LSF trajectory behavior across the converter system. A low pass filter is incorporated in the
consecutive speech frames." circuit to filter out unwanted harmonics and to give a
sinusoidal AC current. A high frequency transformer with
1:1 ratio is incorporated in the design to provide galvanic
isolation for better circuit performance and protection. The
optimum operation of transformer core in four quadrant of
B-H curve is also considered in the proposed converter
topology. The laboratory model of the converter is
developed and tested. The experimental result is compared
with the simulation result.

Tencon 2005 Page 64

 1D-12
Session Title: Power Generation
Chairperson: Amin Sadik
Day: Tuesday
Date: 22nd November 2005
Time: 4:10PM – 5:50PM
Room: M9

1D-12.1 1568964506 1D-12.3 1568964783

Generation Expansion Planning Including Assessments for the Impacts and Benefits of
Biomass Energy Sources with Wind Farm Placement
Dulyatat Nualhong, Songsak Chusanapiputt, Sujate Jen-Hao Teng, Chin-Ling Yu
Jantarang, Vuthichai Pungprasert "Wind generation plays a major role on renewable energy
"This paper presents a development of Improved Tabu sources; it is very important for future energy developments
Search (ITS) and its application to a least-cost Generation and renewable energy industries. Therefore, the power
Expansion Planning (GEP) including biomass energy system impacts and economic benefits of wind generation
sources under global environmental impact consideration. A must be clarified before wind farm placement and
carbon tax is taken into account as countermeasure to installation. In this paper, a simple assessment method for
reduce CO2 emission for introducing and promoting wind farm placement combining the power system impact
biomass energies to GEP problem based conventional assessment and economic benefit analysis is proposed. For
fossil-fuel plants. The proposed ITS is conducted by the a candidate wind farm location, the maximum permitted
proposed a self-reforming candidate list strategies to wind generation capacity can be estimated by analytical
improve search performance of a standard TS. The ITS formulas with respect to the interconnection requirements.
approach is applied to the test system with 15 existing After the maximum permitted capacity was obtained, a
power plants, 5 types of conventional fossil-fuel and 2 types benefit/cost analysis can be used to determine whether this
of biomass energy over a 13-year planning period. The investment project is profitable. In order to simplify the
simulation results reveal that the proposed ITS not only can whole planning procedures, a user-friendly Human-Machine
yield optimal solution as well as dynamic programming, but Interface (HMI) used for data input, interconnection
also can remarkably reduce computational time. Moreover, requirement selection and final report printout is also
the proposed method provides better solution superior to a designed in this paper. Due to the uncertainty
standard tabu search with promising results." characteristics of wind power, the sensitivity analysis for
economic benefits with respect to the variations of wind
1D-12.2 1568964510 speed and Wind Turbine (WT) cost etc. are also integrated
into the proposed HMI. The assessment methods
Relative Velocity Updating in Parallel Particle developed in this paper will be helpful for wind power
Swarm Optimization Based industries. Test results demonstrate the validity of the
proposed assessment procedures."
Dulyatat Nualhong, Songsak Chusanapiputt, Sujate
Jantarang, Sukumvit Phoomvuthisarn
1D-12.4 1568965418
"This paper presents an effectiveness of combined Parallel
Relative Particle Swarm Optimization (PRPSO) and Energy-saving Flyback Converter for EDM
Lagrangian Relaxation (LR) for a large-scale constrained Applications
Unit Commitment (UC) problem in electric power system. Carl Michael Odulio, Luis Sison, Miguel Escoto
The proposed algorithm incorporates PRPSO with a new
Relative Velocity Updating (RVU) approach to tradeoff the "Electrical discharge machining (EDM) uses controlled
solution of each slave processing unit. The parallel electric sparks to erode the metal in a workpiece, and this
algorithm based on the synchronous parallel method is now a well established machining process for
implementation is developed to consider the neighborhoods high power applications. EDM power supplies utilizing
decomposition of multiple particle swarm optimizers. The different topologies for high power applications (e.g. full
proposed PRPSO divides the neighborhood into sub- bridge configuration) are widely available. Through the
neighborhood so that computational effort is reduced and years, EDM processes have increasingly been used in high
UC solutions are remarkably improved. The proposed precision machining and in manufacturing micromechanical
method is performed on a test system up to 100 generating components, with new low power topologies being
units with a scheduling time horizon of 24 hours. The investigated for the latter. We implemented a current mode
numerical results show an economical saving in the total flyback converter for use in low power EDM applications.
operating cost when compared to the previous literature This topology has an energy conservation feature and
results. Moreover, the proposed PRPSO based RVU removes the need for output bulk capacitors. Energy used
scheme can considerably speed up the computation time of in the erosion process will come from the energy stored in
a traditional PSO, which is favorable for a large-scale UC the flyback transformer and will be transferred to the small
problem implementation." output filter capacitors and the effective capacitance of the
tool, workpiece, and dielectric combination. A
servomechanism that controls the arc distance will
determine the output voltage. We have developed an EDM
power supply prototype utilizing the flyback topology for low
power applications with minimal components count and
inherent protection under short circuit conditions. Our work
includes the design, fabrication, and characterization of the
whole system. We compared the efficiency and quality of
work of our proposed topology with that of a linear
converter."

Tencon 2005 Page 65

 2B-01
Session Title: Network Performance and Design 1
Chairperson: Irena Atov
Day: Wednesday
Date: 23rd November 2005
Time: 10:50AM – 12:30PM
Room: M9

2B-01.1 1568964098 2B-01.4 1568965096

"Power, Rate and QoS Control for Impulse Radio" A Failure Analysis of the Tomogravity and EM
Rudolf Mathar, Anke Feiten Methods
"Because of its limitations due to interference, ultra- Suyong Eum, Richard Harris, John Murphy
wideband impulse radio needs a careful investigation of the "A traffic matrix provides a major input to the design,
signal-to-interference ratio (SIR). This paper contributes to planning and management of a telecommunications
understanding the theoretical foundations of the SIR effects network. As the Internet is being proposed as the principal
in multi-user impulse radio systems. Two basic aspects are delivery mechanism for telecommunications traffic at the
investigated. First, the set of admissible power allocations present time and this network is not owned or managed by
to maintain certain quality-of-service thresholds is a single entity, there are significant challenges for network
described. Secondly, the geometry of the set of feasible planners and managers needing to determine equipment
reciprocal bit rates is characterized. Both sets turn out to and topology configurations for the various sections of the
have certain monotonicity and convexity properties. We Internet that are currently the responsibility of ISPs and
furthermore discuss the case that the pulse repetition time traditional telcos. Planning of these subnetworks typically
is used as a global parameter to achieve an admissible requires a traffic matrix of demands that is then used to
power allocation at the price of proportionally reduced infer the flows on the administrator's network. Unfortunately,
quality-of-service parameters. Techniques from analyzing computation of the traffic matrix from measurements of
code division multiple access systems are generalized and individual flows is extremely difficult due to the fact that the
applied to impulse radio in this paper." problem formulation generally leads to the need to solve an
under-determined system of equations. Thus, there has
2B-01.2 1568964734 been a major effort from among researchers to obtain the
traffic matrix using various inference techniques. In this
On the Periodicity of Time-series Network and paper, we have studied the impact of the underlying
Service Metrics assumptions for two methods that have shown promise in
Joseph Lizier, Terry Dawson the estimation of Internet traffic demand matrices known as
the Tomogravity and EM (Expectation and Maximization)
"The presence of an underlying periodicity in time-series methods respectively. As the Tomogravity model is a
network and service metrics has been used as a basis for combination of the well-known gravity model and the
some recent anomaly detection techniques. These method of least squares we have also considered the
techniques however assume the presence of a periodicity, problem of obtaining a good prior traffic matrix for the least
and would benefit from the concept of a quantitative figure squares component of this model. We have demonstrated
of merit for the strength of a given periodicity in the metric. that the accuracy of these methods is highly dependent
We survey a number of potential techniques for this upon the underlying assumptions of these models, as well
purpose, and find none suitable. As such, we construct as the selection of an appropriate prior traffic matrix."
such a figure of merit to suit our application. Use of the
figure of merit allows selection of the most appropriate 2B-01.5 1568964949
period for the metric, and we present an efficient automated
method for this selection. Furthermore, this figure of merit is Establishing Physical Survivability of Large
a useful indicator of whether periodic analysis for anomaly Networks using Properties of Two-Connected
detection is in fact suitable for the given metric. Finally, we Graphs
suggest a number of other areas where use of the figure of
merit could enhance anomaly detection using periodic Daryoush Habibi, Hoang Nghia Nguyen, Quoc Viet
analysis." Phung, Kung-meng Lo
"Establishing the physical survivability of large networks is
2B-01.3 1568964948 not a trivial task. Some techniques for assessing physical
Design of an Enhanced Fairness Algorithm for survivability such as the cutset method can not deal with
the IEEE 802.17 Resilient Packet Ring large size networks [1], [2]. A fast technique for finding
biconnected components of a graph and testing the network
Tae Kim for node-/link-bridges, presented in [3], does not provide
"The fairness algorithm of the Resilient Packet Ring IEEE any further information, such as identifying the fundamental
802.17 standard suffers from throughput degradation under cycles within the network, which would significantly benefit
an unbalanced overload. This paper designs an enhanced the next phase of network design for protection using such
aggressive mode of fairness algorithm to resolve this techniques as shared backup path protection (SBPP), p-
throughput degradation and discusses its performance. cycle, or ring protection [3]. This paper presents an
Under the enhanced aggressive mode, upstream nodes set alternative technique, based on graph theory, for evaluating
their allow_rates as their maximum values when a the physical survivability of networks. This technique can
congested node advertises congestion alleviation to them. deal with network sizes of many thousand nodes, with
The congested node decides its congestion alleviation once computational times which are comparable with the
its STQ length goes below a threshold. A new fairness biconnected components method, whilst providing more
packet of type 3 is used for the congestion alleviation information about the susceptibility of a network to
advertisement." individual link and node failures in preparation for the next
phase of network protection design."

Tencon 2005 Page 66

 2B-02
Session Title: Optical Networks
Chairperson: Malin Premaratne
Day: Wednesday
Date: 23rd November 2005
Time: 10:50AM – 12:30PM
Room: M7

2B-02.1 1568965688 2B-02.4 1568964872

Optical Monitoring Issues in Highly Transparent "Alternative Architectures for Bidirectional Single
Optical Networks Mode Fiber SuperPON 512 ONU, 100 km"
Kerry Hinton, Sarah Dods, Peter Farrell, Don Hewitt, Sakena Abdul Jabar
George Dhosi, Trevor Anderson This paper presents alternative architectures for high split
"An overview of the applicability of current optical network and long range bidirectional single mode fiber (SMF) super
monitoring technologies is given. It is shown that these passive optical network (SuperPON). The erbium doped
technologies are not well suited to fault analysis and fiber amplifier (EDFA) is introduced to enhance power
location in the next generation of highly transparent budget. Three different alternative architectures are studied:
dynamically switched optical networks. A new monitoring 1) using transmission speed of 10 Gbit/s and 2.5 Gbit/s at
method, based on heterodyne detection is also described. same wavelength of 1550 nm; 2) using same transmission
This method enables information about the phase of the speed of 10 Gbit/s at wavelength of 1550 nm for
optical field to be extracted." downstream and 1530 nm for upstream; 3) using
bidirectional EDFA at transmission speed of 10 Gbit/s and
2B-02.2 1568963590 2.5 Gbit/s with same wavelength of 1550 nm. The
elaboration of architectures together with their advantages
Telstra's Fibre To The Premises Network Trial and disadvantages are included. The feasibility of deploying
Bruce Clarke a SuperPON with 512 optical network units (ONUs) at
transmission length of 100 km was observed from power
"Telstra is conducting a field trial of a Fibre To The budget studies.
Premises (FTTP) network delivering the triple play of
services, voice, high speed data and video to residential 2B-02.5 1568965149
customers. This paper describes the Broadband Passive
Optical Network architecture employed in these Cost Model based Configuration Management
Queensland trials. An assessment of the FTTP voice quality Policy in OBS Networks
indicates that performance should at least match the PSTN
for the majority of calls. This paper also describes design Hyewon Song, Sang-Il Lee, Chan-Hyun Youn
aspects of a cost-effective scalable video network which "In OBS networks, the one-way reservation strategy causes
can support up to 800 MPEG2 channels on 80 64-QAM the blocking problem by the contention in a resource
channels." reservation. In order to solve this problem, we propose the
configuration management policy based on operation cost
2B-02.3 1568964770 model. We develop the cost model based on DEB
according to the network status information changed by
Scalable AWG-based Multi-Protocol Optical guaranteed QoS, and develop the criteria for the
Switch (MPOS) for Future Optical GRID configuration management by providing the alternate path
Networking using bounded range of the sensitivity of this cost. Finally,
Bernard Lee, Mohamad Romli, Kaharudin Dimyati throughout the theoretical and experimental analysis, we
can obtain that our proposed scheme has good
"This paper presents a novel optical switch, MPOS concept performance in an aspect of blocking rate and outperforms
which is capable of handling circuit, burst and packet based the conventional scheme in complexity as results."
traffic and also perform all optical multicasting. This switch
design is based upon a wavelength routed switching matrix
with the used of SOAs as wavelength converters and also
to perform multicasting. Optical Switches with such
capabilities are essential for future Optical Grid Networking.
Physical performance of the MPOS at 10Gbps will be
presented in this paper."

Tencon 2005 Page 67

 2B-03
Session Title: Routing and Mobile IP
Chairperson: Sanjay Bose
Day: Wednesday
Date: 23rd November 2005
Time: 10:50AM – 12:30PM
Room: P2

2B-03.1 1568964747 2B-03.3 1568965285

Improved Vector Route Algorithm Bases on Design of Multicast Routing and Wavelength
Baster Topological Space Model Assignment in Multifiber WDM Mesh Networks
Guo Min Gu, Weihong Wang for Asymmetric Traffics
“Net analysis is one of the most important space analysis Warangrat Wattanavarakul, Sak Segkhoonthod,
functions. Finding a route is the basic function of net Lunchakorn Wuttisittikulkij
analysis in GIS, the kernel of which is to get the optimized "This paper proposes new mathematical formulations for
route. This paper presents a topological space model, solving the multicast routing and wavelength assignment
which saves all points in the vector-graph layer based on a (MC-RWA) in multifiber WDM mesh networks for
grid method and improves the classical Dijkstra algorithm asymmetric traffics. We investigated MC-RWA problems in
bases on the raster topological space model. This improve three different MC-RWA network design approaches
efficiency in time of the route planning. At the same time, it namely Light-Tree (LT), Partial Virtual Light-Tree (PVLT),
decreases the memory occupancy, so the new algorithm and Virtual Light-Tree (VLT) meshed network. Given a
can be used on a small memory terminal. Keywords: static multicast asymmetric traffic set, the number of
optimum path analysis; raster; space analysis; Dijkstra wavelengths a fiber can support, and the network design
algorithm" approach (LT, PVLT, or VLT), we seek to realize the
minimal number of fiber requirements. Numerical results on
2B-03.2 1568965275 a small test network are given for highlighting the key
Dynamic Router Tables for Full characteristics of each MC-RWA strategies. Based on the
test network, the numerical results demonstrate that in most
Expansion/Compression IP Lookup cases PVLT and VLT networks require fewer fibers than LT
Sieteng Soh, Lely Hiryanto, Suresh Rai, Raj P. network, signifying that wavelength conversions can be
Gopalan useful for reducing fiber requirements in asymmetrical
multicast environments using mesh design approaches.
"We propose a dynamic-Full-Expansion-Compression Nonetheless, conversion is not always needed, as partial
(DFEC) technique that requires exactly three memory wavelength converter allocation appears to be as effective
accesses for each IP address lookup as that required in the as full allocation with respect to capacity requirement under
static-FEC. A prefix update in the static scheme needs table certain network environments."
reconstruction that requires hundreds of milliseconds, and
hence the approach is unsuitable for use in a dynamic- 2B-03.4 1568963448
environment. DFEC selectively decompresses the table to
obtain addresses that are affected by a prefix update, Fast MPLS Network Optimisation using Machine
modifies their next-hops, and recompresses the data Learning
structure. The resulting tables will be identical to what would
be obtained by a full reconstruction, but at much lower Michael Dale, Herman Ferra, Robert Palmer
computational cost. We have implemented our algorithm in "This paper presents the application of a machine learning
ANSI-C and have run it on a 3.2 GHz Pentium-4 to build an technique (Neural Networks) to the problem of optimising
FEC table of each of the following five databases: the traffic flows (Label Switched Path placement) within
AADS2001, Mae-West2001, Mae-East1997, Paix2001, and telecommunications networks using Multi-Protocol Label
PB2001. Our simulations show that an update in DFEC on Switching (MPLS). It is shown here that the machine
each of the tables takes less than 7 microseconds." learning technique achieves very fast computation of
accurate solutions to the problem of placing Label Switched
Paths in order to maximise the performance of the network.
Results are presented which compare the accuracy of the
Neural Network solutions with the optimal solutions
produced by the standard Mixed Integer Linear
Programming technique, and the time taken to produce the
solutions. As well, network simulation results are presented
which show the improvement in network performance, in
terms of packet loss, that can be achieved when using such
a technique."

Tencon 2005 Page 68

 2B-04
Session Title: Image Processing 3
Chairperson: Henry Wu
Day: Wednesday
Date: 23rd November 2005
Time: 10:50AM – 12:30PM
Room: P1

2B-04.1 1568964829 were enhanced effectively by using the compensated

results."
Bias Characteristics of Bilinear Interpolation
Based Registration 2B-04.4 1568965266
Donald Bailey, Andrew Gilman, Roger Browne Novel Masks for Multimodality Image Fusion
"Image registration has application in many image using DTCWT
processing tasks. A new gradient-based registration method
that uses the bilinear interpolation equation to predict the
Mohd. Shahid, Sumana Gupta
pixel values of one image from those in a reference image "We present an image fusion technique using the Dual tree
is developed and its bias characteristics analysed. For complex wavelet transform (DT-CWT). We have proposed
sinusoidal signals and step edges, the estimated offset is novel masks to extract information from the decomposed
biased towards the nearest reference pixel, however for structure using DT-CWT. The main goal of this paper is to
noise signals, the bias is away from the nearest pixel. introduce a new approach to fuse multimodality images
Therefore the bias (and registration accuracy) can actually using dual tree complex wavelet transform. Experiment
improve with the addition of noise. This phenomenon is results show that the proposed fusion method based on
demonstrated with two sample images with quite different complex wavelet transform is remarkably better than the
image statistics." fusion method based on classical discrete wavelet
transform. This method is relevant to visual sensitivity and
-04.2 1568965182 tested by merging multisensor, multispectral and
defoucused images apart from CT and MR images. Fusion
A Mobile Motion Capture System Employing is achieved through the formation of a fused pyramid using
Image Transfer the DTCWT coefficients from the decomposed pyramids of
Iku Yamaguchi, Joo Kooi Tan, Seiji Ishikawa the source images. The fused image is obtained through
conventional inverse dual tree complex wavelet transform
"This paper proposes a technique for recovering 3-D shape (DTCWT) reconstruction process. Results obtained using
of remote non-rigid objects employing mobile cameras. the proposed method show a significant reduction of
Since the proposed technique allows independent move of distortion."
the employed cameras, 3-D recovery under various
environments such as remote places can be realized. Our 2B-04.5 1568965308
technique has an advantage over others in that camera
calibration before taking images is not prerequisite to the Robust Background Subtraction based on Bi-
recovery. A system is described that captures images of polar Radial Reach Correlation
remote objects by the proposed motion capture technique,
transfers the images by analog airwaves, and recovers 3-D
Yutaka Satoh, Katsuhiko Sakaue
shape of the objects from the images. It is expected that "Background subtraction algorithms are widely utilized as a
this technique will improve the efficiency of image technology for segmentation of background and target
information transfer. In the performed experiment, a human objects in images. In particular, the simple background
walking motion captured at a remote place recovered its 3- subtraction algorithm is used in many systems for its ease
D motion successfully at a laboratory by image transfer." and low cost of implementation. However, because this
algorithm relies only on the intensity difference, it has
2B-04.3 1568965229 various problems, such as low tolerance for poor
illumination and shadows and the inability to distinguish
A Similarity Analysis of DGGE Images using DP objects from their background when their intensities are
matching similar. In an earlier study we proposed a new statistic,
Yuka Sugiyama known as Radial Reach Correlation (RRC), for
distinguishing similar areas and dissimilar areas when
"In this paper we propose a new method of similarity comparing background images and target images at the
appraisal between samples in the image obtained by pixel level. And we achieved a robust background
polymorphism analysis called DGGE (Denaturing Gradient subtraction by evaluating the local texture in images. In this
Gel Electrophoresis). DGGE uses an electrophoresis unit study we extended this method further and developed a
and needs to observe reactions of samples included in the method to ensure stable background separation even in
gel. Therefore the samples are often visually distorted. cases where the image texture is feeble and the intensity
Conventionally the analysis of the DGGE images has been distribution is biased."
performed as visual evaluations, which requires a lot of
human labor. Moreover the reliability of the analysis highly
depends on skills of the operator and the reproducibility the
burden of human labor and various others. Then, the
process of extraction of data corresponding to samples from
a DGGE image was automated, and a method of pattern
matching with a degree of similarity for each sample was
introduced. A distortion of DGGE was compensated by use
of a DP matching, often used in the signal processing. A
degree of similarity and classification between samples

Tencon 2005 Page 69

 2B-05
Session Title: IT Applications 3
Chairperson: Horace King
Day: Wednesday
Date: 23rd November 2005
Time: 10:50AM – 12:30PM
Room: P3

2B-05.1 1568965239 2B-05.3 1568965333

Vision Based Motion Tracking System for Topologically Robust Surface Reconstruction for
Interactive Entertainment Applications 3D Shape Modification
Jaeyong Chung Junta Doi, Wataru Sato
"Although the vision based motion recognition plays an "This paper proposes a topologically robust and accurate
important role in virtual reality and interactive entertainment shape modeling procedure that approximates a real 3D
area, it has not yet found wide usage. The first main reason object, based not on the widely used so-called ICP (Iterative
is the unavailability of relatively low cost wireless tracking Closed Point) algorithm, for the noise resistive modeling
system that would operate well enough for inputting and especially for the organized and matrix-formed
gestures. The second reason lies in the lack of naturalness meshing, enabling the quantitative 3D shape modification
in gesture design. That is, so far most gestures used are easy. A geometric model with desired meshing is directly
either static images or poses and too abstract losing its reconstructed based on a solid modeling approach. The
intended meaning and affordances, resulting in low usability radial distance of each scanning point from the axis of the
and presence. In this paper, we propose architecture for a cylindrical coordinates is measured using a laser
low cost re-configurable vision-based motion tracking triangulation sensor. The angular and vertical positions of
system and motion recognition algorithm. The user wears the laser beam are two other coordinate values of the
one or more retro-reflective markers for achieving more scanning to be simultaneously acquired. A face array listing
exact tracking performance and make predefined motion (topology), which defines the vertex (sampling point)
gestures. Through object segmentation processing , 3D connectivity and the shape of the mesh, is assigned to meet
positions of the objects are computed. And then, a motion the desired meshing. Some bioengineering examples are
gesture corresponding on these 3D position trajectories is demonstrated for the quantitative shape modification."
found by a simple correlation-based matching algorithm.
Also, we demonstrate this system by applying it to virtual 2B-05.4 1568965374
environment navigation and interactive entertainments."
End-users resource consumption of spam and a
2B-05.2 1568965287 3D anti-spam evaluation framework
Transient Stability Program Using Component- Minh Tran, Grenville Armitage
Based Software Engineering "Our work is motivated by the challenge of coupling
excessive network consumption and control of the
Hadi Suyono, Khalid Mohamed Nor, Sallehhudin consumption; the importance of cost models to quantify and
Yusof characterize the consumption; and the crucial requirements
"This paper presents the development of a transient stability of controlling techniques to limit the theft of network
analysis (TSA) program using component based software resources. In this paper, we particularly examine the
engineering (CBSE). The TSA requires the integration of network consumption of spam. We quantify the end-users
many analyses such as linear solver (LS), load flow network resource consumption caused by spam as a
analysis (LFA), and numerical integration methods. The function of different email retrieving mechanisms, and
CBSE allows existing components to be reused and estimate the financial costs of spam for both residential and
integrated with newly developed components. Therefore, business users. The paper also describes our anti-spam
the TSA program can be developed by reusing an existing rate-limiting tool, MT Proxy, and introduces a general three-
LS and LFA software components. Only software dimensional framework to evaluate anti-spam techniques.
components for numerical integration methods are required We examine and compare both traditional and novel anti-
to be developed from scratch. The TSA and LFA software spam solutions and show a simple comparison using our
components are developed based on object oriented design model."
(OOD). This design involves modeling the power system
devices as entity objects based on the electrical network
primitive data. The advantages of the proposed TSA
application are speeding up programming and saving the
development resources. In addition to, any component can
be replaced or updated without altering or recompiling the
other components in the TSA application. The results in
terms of execution time show that the proposed TSA
application has performance similar to the conventional
procedural development."

Tencon 2005 Page 70

2B-05.5 1568965380
Anti-collision algorithm using Bin slot in RFID
System
Ho-Seung Choi, Jae-Hyun Kim
"Anti-collision algorithm is very important in the RFID
system, because it decides tag identification time and tag
identification accuracy. We propose improved anti-collision
algorithms using Bin slot in RFID system. In the proposed
algorithms, if the reader memorizes the Bin slot information,
it can reduce the repetition of unnecessary PingID
command and the time to identify tags. If we also use
ScrollAllID command in the proposed algorithm, we know
the sequence of collided ID bits. Using this sequence, we
can reduce the repetition of PingID command and tag
identification time. We analyze the performance of the
proposed anti-collision algorithms and compare the
performance of the proposed algorithms with that of the
conventional algorithm. We also validated analytic results
using simulation. According to the analysis, for the random
tag ID, comparing the proposed algorithms with the
conventional algorithm, the performance of the proposed
algorithms is about 130% higher when the number of the
tags is 200. And for the sequential tag ID, the performance
of the conventional algorithm decreases. On the contrary,
the performance of the proposed algorithm using ScrollAllID
command is about 16% higher than the case of using
random tag ID."

Tencon 2005 Page 71

 2B-06
Session Title: Power Quality Monitoring 1
Chairperson: Changliang Liu
Day: Wednesday
Date: 23rd November 2005
Time: 10:50AM – 12:30PM
Room: M3

2B-06.1 1568960541 2B-06.4 1568964633

Bayes Method of Power Quality Disturbance Embedded Web Server Based Power Quality
Classification Recorder
Jidong Wang, Chengshan Wang Yu-Hung Chen, Ming-Jieh Cheng, Lain-Chyr Hwang,
"With the proliferation of nonlinear loads, power quality Jen-Hao Teng, Shun-Yu Chan
problems have been paid more attention to. In order to "An Embedded Web Server (EWS) based Power Quality
mitigate the influence, various power quality disturbances (PQ) recorder is designed in this paper to achieve the goal
must be classified before an appropriate action can be of PQ remote monitoring and recording. An EWS is a Web
taken. Wavelet packet is developed on wavelet transform, server that runs on an embedded system and is possible to
which can provide more plenteous time-frequency provide a powerful Web-based management and control
information. This paper selects energy and entropy of interface with tiny volume. Therefore, it will be very suitable
terminal nodes through wavelet packet decomposition as for PQ remote monitoring and recording. The experimental
feature vector respectively, using Bayes classifier to classify platform used in this paper is a Samsung S3C4510-based
the disturbances, which are simulated and analyzed. The embedded system. This embedded system features an
simulation results indicate that the entropy acted as feature uCLinux operating system; therefore, the design and
vector has higher recognition accurate ratio." implementation of the proposed EWS system will be more
efficient and effective. An embedded PQ analyzer with
2B-06.2 1568963632 appropriately designed I/O interface and peripherals is
designed and implemented firstly. The PQ analyzing
COMTRADE-Based Fault Information System for functions implemented in the proposed system include 1)
TNB Substations system voltage and frequency calculation, 2) harmonic
Gan Chin Kim, Abdullah Asuhaimi Mohd Zin, spectra and total harmonic distortion calculation and 3)
Zainoren Shukri abnormal event recorder etc. Then, a Web Human-Machine
Interface (HMI) programmed using ActiveX technology,
"This paper describes the process of extracting the fault which is suitable for EWS and can be used for PQ
information from Common Format for Transient Data monitoring, is designed. The proposed EWS based PQ
Exchange (COMTRADE) record. The COMTRADE format recorder can be used for remote monitoring with minimum
record was obtained from the distance relay recording size and maximum efficiency. It will be also helpful for large-
device, provided by the Protection Department of Tenaga scale and wide-area PQ monitoring and recording.
Nasional Berhad (TNB). A graphical user interface (GUI) Experimental results demonstrate the validity of the
simulation program using MATLAB was developed which proposed system."
implements the one-cycle cosine filter relaying algorithm to
digitally filter out the unwanted signals at systems 2B-06.5 1568964816
fundamental frequency from the COMTRADE record. This
work aims to help the TNB protection engineer to have a Three Phase Generalised Power Theory for the
faster solution in parameter extraction from protective relay Measurement of Powers in an Electric Power
recorded from substations in COMTRADE format." System

2B-06.3 1568963708 Harnaak Khalsa, Jingxin Zhang

"The deficiencies of present power measurement definitions
Structure Design For Real-Time Thermal and techniques in the presence of harmonics is a known
Performance Analysis And Monitoring System Of problem. As a solution to this problem, a new Three Phase
Power Plant Power Theory (3PT), which splits the instantaneous power
Changliang Liu into active and non-active instantaneous power, is
proposed. With good knowledge of the active/non-active
"The real-time thermal performance analysis and monitoring power components an accurate measurement especially of
system (RTPAMS) of power plant is the guarantee that the non-active component is achieved. Also this non-active
power plants work in economic operating condition.Four component can be utilised to achieve better compensation.
different structures of the system are summarized: This paper presents the new 3PT for the three-phase
Independent- RTPAMS,RTPAMS basing on DCS,RTPAMS system. The performance of the 3PT for linear and
basing on SIS,RTPAMS basing on MIS. It also analyzes nonlinear loads in the presence of harmonics is illustrated
characters and acclimatization of them. Finally, it forecasts with case studies. The case study uses unbalanced
the trend of RTPAMS and puts forward a new scheme of resistive loads with diode to create nonlinear load. Results
RTPAMS, basing on demands of electric information of case studies and applications and analysis of the results
system." demonstrate good performance of the proposed 3PT."

Tencon 2005 Page 72

 2B-07
Session Title: Adaptive Systems and Signal Processing 1
Chairperson: Kevin Lin
Day: Wednesday
Date: 23rd November 2005
Time: 10:50AM – 12:30PM
Room: M6

2B-07.1 1568964062 2B-07.4 1568964904

A Geometric Mean Based Adaptive Local Noise Improved Performance for Stereophonic
Removal Algorithm Acoustic Echo Cancellation using Subband
Sun QiaoPing, Zhao Xiaoming Structure
"Adaptive filters are important application in the signal Boonchai Krittayanun, Nisachon Tangsangiumvisai
processing field. This paper discusses the adaptive local "An adaptive filtering approach is proposed in this paper for
noise removal algorithm proposed by Professor Rafael C. Stereophonic Acoustic Echo Cancellation (SAEC) in order
Gonzalez and points out its shortcomings. To improve the to obtain convergence rate improvement and computational
algorithm, a new method, geometric mean based adaptive complexity reduction. The hybrid mono/stereo structure
local noise removal algorithm, has been proposed. The between Fast Affine Projection (FAP) and Frequency-
simulation results indicate that the new algorithm is more domain Normalized Least Mean Square (F-NLMS)
satisfactory. The mean square error is reduced by 1/4. The algorithms is introduced, based on subband adaptive
signals to noise ratios (i.e., SNR, SNRm, PSRN) are raised filtering. Simulation results on speech signals suggest that
by 1/10.This algorithm has been shown promise for enhanced performance in terms of rate of convergence can
applications." be achieved at low computational cost."

2B-07.2 1568964704 2B-07.5 1568964930

A Linear Fusion of the Local Kalman Estimates Performance of Adaptive Transmit Diversity with
Vladimir Shin, Rashid Minhas, Georgy Shevlyakov, Orthogonal Space-Time Block Coding in
Kiseon Kim Microcell and Macrocell Channel Environments
We extend our recently obtained fusion formula yielding an Seedahmed Mahmoud, Zahir Hussain, Kevin Lin
optimal mean square combination of unbiased estimates on "Recently an adaptive transmit eigenbeamforming with
the case of an arbitrary number of biased estimates and orthogonal space-time block coding (Eigen-OSTBC) has
apply it to fusion of multisensor local estimates. The been proposed. This model was simulated over macrocell
suboptimal two-stage filter for linear dynamic systems is environment with a uniform linear array (ULA) at the base
designed: the local optimal Kalman estimates computed at station (BS). In this paper we investigate the impact of
the first stage are linearly fused at the second stage. Its various antenna geometries on the performance of this
realization needs a lower memory demand than the optimal scheme as well as the performance over microcell and
Kalman filter. The example exhibits the effect of noise on macrocell channel environments. The geometrical-based
the performance of fusion of the state estimates based on hyperbolically distributed scatterers (GBHDS) channel
measurements from different sensors. models were simulated with angular spreads (AS) taken
from experimental data. The ULA and the uniform circular
2B-07.3 1568964765 array (UCA) are considered at the BS. In the simulation
Discrete-time Smoothing Formulas conducted, results have shown that the Eigen-OSTBC
system has a higher performance gain than OSTBC
Garry Einicke systems in macrocell than microcell environment (where the
The paper presents formulas for the discrete-time minimum- AS gets larger). It is also observed that BS antenna array
variance fixed-interval smoother. The solution involves a with different separation distances also affects the
cascade of a Kalman predictor and an adjoint Kalman performance of both Eigen-OSTBC and OSTBC systems.
predictor. Speech enhancement and nonlinear However, the error-rate performance curves of ULA and
demodulation examples are presented which demonstrate UCA configurations are very similar."
that linear and extended Kalman smoothers can provide
performance benefits.

Tencon 2005 Page 73

 2B-08
Session Title: Signal Processing Hardware 1
Chairperson: Qassim Nasir
Day: Wednesday
Date: 23rd November 2005
Time: 10:50AM – 12:30PM
Room: M4

2B-08.1 1568963693 behaviour and verified by mixed signal VHDL-AMS

simulations."
A scalable parallel computational core for
embedded processing 2B-08.3 1568965067
Refik Shadich, Ian McLoughlin A Study on Constructing the Arithmetic Unit over
"Embedded computational hardware has become prevalent GF(3m)
in recent years for communications signal processing for
reasons including size and cost. The availability of
Chun Park
competing single processor solutions from traditional "In this paper, we presented a method of constructing a
vendors gives system designers a degree of choice. Some multiplier over finite field GF(3m). We proposed the
recent market entrants have even embraced parallel multiplication arithmetic algorithm and mod F(X) arithmetic
concepts within their architectures. However the fact at the same time. The proposed multiplier is composed as
remains that while one particular computational device or following parts. There are multiplication part, data
parallel configuration may suit a given application, it seldom assortment part, Pj term accumulation and sequential
suits a broad range of other applications. This promotes selection part, descending order mod F(X) generation part
design inefficiency: either developers familiar with one and multiplication processing part. Especially, the proposed
solution from a previous project choose to use it for the next multiplier have regularity, expansibility and modularity of
project despite some probable degree of mismatch, or they circuit design."
are faced with a costly learning curve implied in the
adoption of a different, but possibly better matched, 2B-08.4 1568965091
architecture. A preferable approach is to allow
computational hardware to be adapted at a micro- and A Novel Wide-Swing Wide-Bandwidth Scalable
macro-architectural level to fit requirements on a project-to- Low-Voltage Analog CMOS Multiplier for
project basis, but maintaining common instruction set and Communication Signal Processing
development tools. This gives designers the flexibility to
S. M. Hasan
choose the degree of parallelism and type of parallel
arrangement required for their application, but without This paper reports a novel scalable CMOS four-quadrant
requiring a new tool and hardware learning curve. This multiplier design with wide-signal swing and wide bandwidth
paper describes the 2ke, a flexible and modular enabling numerous applications in communication signal
computational system that allows developers to standardise processing. The multiplier operates in a scalable voltage
on one processor, instruction set, software architecture and range between 3.3V and 1V power supply with a maximum
toolchain for many projects. Architectural enhancements to input dynamic range of 500mV(@ 3.3V supply) at both the
its forerunner, the 2k2, are presented to permit micro- inputs (multiplier & multiplicand inputs) and a -3dB
architectural parallelism to be chosen along a continuum bandwidth of around 1GHz. The simulations were carried
from SISD at one extreme to full SIMD at the other, whilst out using the 3M1P 0.5um Agilent (HP) CMOS (with 3.3V
the very nature of the 2ke permits extension to MIMD along supply) and the 5M1P 0.18um TSMC CMOS (with 1V
an orthogonal development direction. Results in terms of Supply) process technologies. The typical THD (Total
logic cell usage, current consumption and memory usage Harmonic Distortion) for the multiplier was around -56dB.
will be presented for each arrangement for example The multiplier was found to dissipate a mere 1mW from a
application code." 1V supply.

2B-08.2 1568965025 2B-08.5 1568965187

Calibration Technique for Sigma-Delta Probability of Interpolation for a Mute Sample
Modulation Loops using Radio-Frequency Phase Interpolative A/D Converter with Horizon-Length
Detection Two
Soeren Sappok, Andreas Neyer, Andre Kruth, Ralf Milan Derpich, Daniel Quevedo, Graham Goodwin
Wunderlich, Stefan Heinen “There are two major issues which underly the design of an
"A novel calibration scheme for predistortion ${\Sigma analog to digital converter; namely ""when to sample?"" and
\Delta}$ PLLs is proposed in this paper. In contrast to ""how to represent the amplitude of each sample?"". In the
present calibration algorithms this technique offers a digital majority of past work, these two aspects have tipically been
representation of the high frequency phase characteristic. treated separately. Recently, we have introduced a novel
The architecture uses minimum chip area by synchronously algorithm called MSIC which uses moving horizon
sampling the data content of an asynchronous divider optimization ideas to determine both when and how to
chain. Using this technique to detect the phase difference sample. Our earlier work gave a heuristic description of the
during a step response allows to determine the real loop algorithm and showed, via simulations, that a performance
gain within 7~${\mu}$s with an accuracy better than 0.1\%. gain was achieved in both the required bit rate and the
Obtaining this, the deviation from the desired loop gain can achieved distortion level. This seemingly paradoxical result
be adjusted by a digitally controlled charge pump in order to is due to the interpolative capabilities inherent in MSIC. The
derive the wanted loop transfer function. This is mandatory goal of the current paper is to give theoretical support to the
for predistortion modulation loops. The architecture is MSIC algorithm. In particular, we provide bounds on the
designed by systemtheoretical calculation of the loop probability that a beneficial interpolation occurs for the
particular case of horizon length 2 with flat, unity error
weighting filter."

Tencon 2005 Page 74

 2B-09
Session Title: Orthogonal Frequency Division Multiplex 1
Chairperson: Jean Armstrong
Day: Wednesday
Date: 23rd November 2005
Time: 10:50AM – 12:30PM
Room: M8

2B-09.1 1568961364 new frequency-time encoding scheme that can be used to

exploit frequency diversity branches for broadband OFDM
Full Rate Concatenated Space-Frequency and systems with only one antenna at the MS. By incorporating
Space-Time OFDM over Naftali Fading Channels this with receive beamforming at the base station (BS), the
Nor Noordin, Borhanuddin Mohd. Ali, Sudhanshy instantaneous signal-to-noise ratio (SNR) is maximized and
the system error-rate performance is then further improved.
Shekhar Jamuar
Numerical results showed that systems employed the
"In this paper we investigate 2 transmit and 1 receive proposed transceiver structure have a 4-dB improvement
(2Tx:1Rx) space-frequency orthogonal frequency division over the conventional space-time coding scheme when two
multiplexing (SF-OFDM) and space-time OFDM (ST- receive antennas are used."
OFDM) concatenated with convolutional channel coding as
a redundancy method to combat channel impairments. We 2B-09.4 1568964986
built our SF and ST-OFDM on top of our coded OFDM
(COFDM) with coding rate of ¼ and constraint length of 10. Performance Comparison of Blind Symbol
We simulate our concatenated SF-OFDM and ST-OFDM Timing Estimation in Cyclic Prefixed OFDM
over a multipath fading channel using Naftali 802.11 Systems
channel model under different delay spreads in relation to
outdoor environment. Concatenated SF-OFDM performs
Huimin(Grace) Guo
slightly better than ST-OFDM at delay spread of 2.0 "In this paper, an estimated weighting factor and two
microseconds, but both are almost comparable at all other symbol timing estimators are proposed for timing
delay spreads. These two schemes outperforms other synchronization of cyclic prefixed OFDM systems. The
schemes such as single antenna coded OFDM and performance of the proposed estimators is assessed
Alamoutis Space-Time Block Code (STBC) under these through computer simulations, and compared with that of
conditions by as much as 10 dB at BER of 10-4." the van de Beeks estimator and M¨ ullers estimator.
Experiments demonstrate that one of the proposed
2B-09.2 1568963797 estimators outperforms M¨ ullers and yields a similar
performance van de Beeks but without requiring the priori
Designing for Low ISI in an OFDM Modem knowledge of signal and noise powers."
Scott Leyonhjelm, Mike Faulkner
2B-09.5 1568965131
"One problem that arises in OFDM systems is when the
currently received OFDM symbol is distorted by the Transmission of Compressed Multimedia Data
adjacently transmitted OFDM symbols. This is known as over Wireless Channels Using Space-time OFDM
InterSymbol Interference (ISI) and is a function of the with Adaptive Beamforming
combined impulse response of the wireless channel and the
transmit and receive digital and analog filters. This paper Tasso Athanasiadis, Kevin Lin, Zahir Hussain
illustrates the effect of the digital and analog filters on ISI in "The transmission of multimedia data over wireless
a typical 802.11a/g WLAN OFDM modem. It is shown that channels poses significant constraints on the
the filters, rather than the actual wireless channel, dominate communication system bandwidth, energy, and latency. To
the impulse response. Two techniques, increasing the overcome these bottlenecks to wireless multimedia
digital interpolation ratio and the use of frequency domain communication, various channel coding and transmit
interpolation, are then introduced with the effect of reducing diversity schemes have been proposed. In previous work,
the impulse response and ISI. The result of implementing we have shown that space-time block-coding (STBC) with
these two techniques at the transmitter is to reduce the adaptive beamforming (STBC-OFDM-AB) is an effective
required cyclic prefix by 50% for an ISI requirement of technique for improving the error-rate performance and
40dB." channel capacity of wireless multimedia systems utilizing
OFDM. In this paper, we introduce a transmission system
2B-09.3 1568964828 for multimedia communication employing STBC-OFDM with
Adaptive Receiver Beamforming for Diversity adaptive beamforming incorporating a perceptually-based
image compression coder - which consists of a 2-D discrete
Coded OFDM Systems: Maximum SNR Design wavelet transform (DWT), an adaptive quantizer (with
Kevin Lin, Zahir Hussain thresholding) and variable-length entropy encoding. Initial
simulation results based on the transmission of compressed
"Over the recent years, advance multiantenna transmission
images, showed that the performance improvement
schemes have attracted considerable interest due to their
introduced by STBC-OFDM-AB can be readily observed,
potential benefits in improving the system capacity and
and compared to other transmission methods is better
error-rate performance. As a result, space-time coding and
suited to wireless multimedia communication."
transmit beamforming have emerged as the two most
promising techniques. Because of limited space at the
mobile station (MS) and the fact that download intensive
services are to be introduced in the next generation of
cellular systems, most of research efforts have been
pouring on transmit diversity techniques. In this paper, we
focus on adaptive uplink transmission and reception
techniques for wireless communications and introduce a

Tencon 2005 Page 75

 2B-10
Session Title: Fuzzy Technique Applications 1
Chairperson: Alexander Kist
Day: Wednesday
Date: 23rd November 2005
Time: 10:50AM – 12:30PM
Room: M10

2B-10.1 1568964535 and a fuzzy gain scheduling technique. A fuzzy c-means

clustering technique is used in selecting appropriate
A Fuzzy Adaptive Programming Method of number of membership functions and constructing the
Particle Swarm Optimization shape of membership functions. Then, fuzzy IF-THEN rules
Qi Kang are determined using a fuzzy gain scheduling technique. By
implementation of a fuzzy c-means clustering and a fuzzy
"This paper introduces a novel multi-optimum programming gain scheduling, the need for heuristic method for designing
mode for the particle swarm optimization algorithm. Initially, fuzzy membership functions and rules from expert
to efficiently control the relationship between multi-optimum knowledge is omitted. The proposed fuzzy supervisory PI
information and convergence to the global optimum controller is applied to a pilot plant binary distillation column.
solution, fuzzy logic is introduced into the process of multi- The task of the controller is to maintain the product
optimum distribution state dynamic programming. Then, the composition when the disturbance enters the column in the
fuzzy adaptive programming method based on single- form of the changes in feed flow rate. The results show that
variable and two-dimension-input fuzzy control structure for the fuzzy supervisory PI controllers operate this nonlinear,
optimum and sub-optimum distribution information is multivariable process efficiently."
implemented and simulated with different function
optimization problems." 2B-10.4 1568965382
2B-10.2 1568964955 The Application of Fuzzy Data Mining in Coal-
Fired Boiler Combustion System
Fuzzy Matching Scheme for Stereo Vision based
Electronic Travel Aid Xi-ning Yu, Cheng-lin Niu, Jian-qiang Li
Balakrishnan Ganesan, Sainarayanan "Coal-fired boiler combustion system is a complex multi-
input and multi-output plant with strong nonlinear and large
Gopalakrishnan, Nagarajan Ramachandran, Sazali
time-delay. Currently the research of optimizing the
Yaccob combustion process is monitor oriented instead of decision
"Navigation is a significant barrier to individuals with visual oriented. So it lacks the interaction and decision making
impairments. In response to this barrier, a wearable stereo- with the user. To improve the performance of combustion
vision system was investigated and a prototype system process, the key point is to decide the optimization values
named Stereo Vision based Electronic Travel Aid (SVETA) for the main controllable parameters. This paper proposed a
has been developed. Stereo cameras are employed in the fuzzy data mining algorithm to decide the optimization
SVETA in order to provide information about orientation, values from the history data sets. In order to soften the
distance, shape and size of the obstacles or objects infront partition boundary of the domain, the fuzzy sets theory was
of the user. Stereo vision has long been one of the central introduced into the association mining process. This method
problem in computer vision, and stereo matching is the can translate Quantitative Association Rules Problem
most important and critical issue of stereo vision. In this (QARP) into Boolean Association Rules Problem (BARP).
paper, similarity measure for stereo matching based on Finally some results of instance analysis based on the
fuzzy relations is used to establish the correspondence in practical parameters of 300MW power plant unit are given
the presence of intensity variations in stereo images. The to prove that the fuzzy mining method has good accuracy
strength of relationship of fuzzified data of the windows in and interpretability. The effective results are achieved by
left and the right images of stereo pairs is determined by guiding actual operating based on the optimizing values
considering the appropriate fuzzy aggregation operators. gotten from the fuzzy data mining process."
However, these measures fail to establish correspondence
over the occluded pixels. Left/Right consistence check is 2B-10.5 1568965419
performed to overcome these problems. The final disparity
map is then conveyed to the blind user through musical A Proposed Air-Conditioning System Using
tone based image sonification. Experiments with standard Fuzzy Algorithm for Industrial Application
and real time images have been conducted to demonstrate Md. Shakowat Zaman Sarker
the effectiveness of the algorithms"
"This paper describes the development of a novel air
2B-10.3 1568965197 condition control system based on fuzzy logic. A fuzzy logic
air conditioning controller with two inputs and one output
Implementation of Fuzzy Supervisory PI has been designed with an industrial application in mind.
Controllers for a Binary Distillation Column Using This system consists of two sensors for feedback control:
Fuzzy c-Means Clustering Combined with Fuzzy one to monitor temperature and another one to monitor
Gain Scheduling humidity. There are three control elements: cooling valve,
heating valve, and humidifying valve, to adjust the
KANNAWAT SOMSUNG, Suvalai Pratishthananda temperature and humidity of the air supply. A fuzzy rule for
"Control loops in process control, controlled by conventional this controller is formulated by temperature and humidity.
PID controllers could be improved by the application of The model of this controller has been tested using MATLAB
fuzzy logic-based adapter. However, to obtain an optimal simulation."
set of fuzzy membership functions and rules is not an easy
task. In this paper, a fuzzy supervisory PI controller is
developed based on a fuzzy c-means clustering technique

Tencon 2005 Page 76

 2B-11
Session Title: ATNAC Poster Session 1
Chairperson: Laurie Cahill
Day: Wednesday
Date: 23rd November 2005
Time: 10:50AM – 12:30PM
Room: M1

2B-11.1 1568965166 obtained data about one individual's destinations, average

speed and rest times whilst moving throughout a city-wide
Wireless Ad-Hoc Networking: Analysis and area. We present results from this real life data and useit to
Validation of Simulation Results validate some of the key characteristics of the RWP. In this
Houssein Hallani, Seyed Shahrestani, Frank paper we consider the RWP as a model of user mobility in
networks that cater for a large geographical area - such as
Stootman
a city."
"A wireless Ad-hoc network consists of wireless nodes
communicating without the need for a centralized 2B-11.4 1568961735
administration, in which all nodes potentially contribute to
the routing process. In this paper we report the A hybrid admission control scheme for wireless
experimental results obtained from establishing a small networks
physical network. These results are then compared with Triet Giang, Jidong Wang
those obtained from simulations that are carried out using
OPtimised NETwork (OPNET) software simulator. "Threshold Access Sharing (TAS) is a policy based Call
Correlation between the two sets of results is found to be Admission Control (CAC) scheme in wireless IP networks. It
satisfactory enough to validate the simulation technique and has been shown to adapt well to the limited wireless
process. Given this validation, the investigation of a larger resources. However, its handling of handoff calls is missing.
scale Ad-hoc network is then performed based on similar In this paper, we proposed a hybrid scheme which
simulation techniques. The simulation results of the larger combines an improved TAS scheme with a simplified rate-
scale network confirm our previously obtained results. An based borrowing scheme, aiming to improve its
important aspect of which is that the nature of the drop in performance both on new call blocking and handoff call
throughput with respect to the number of nodes trying to dropping probability. Our analysis and simulation have
communicate simultaneously with a single node, is linear." shown that the new scheme has met the expectation and its
complexity is comparable to TAS."
2B-11.2 1568963136
2B-11.5 1568964993
Internet Protocol Television (IPTV)
Application of Assisted TDOA Technology in
Abdulrahman Yarali, Ann Cherry Vehicle Positioning and Navigation System
"This paper is intended to be an introduction into the issues
involved with delivering IPTV over DSL. IPTV technology
David Lew
and available broadband bandwidths are evolving to the "Based on cellular mobile net, TDOA positioning algorithm
point where DSL providers can effectively enter the market has been demonstrated by many parties in the recent
of TV delivery to their customers. However, there are still years, however, low degree of accuracy compared with
issues concerning bandwidth, codec selection, QoS, and GPS limits its usage in vehicle positioning and navigation
ROI. This paper will address each of these issues and services. Here we introduce two assisted technologies, the
discuss the major players involved in deploying this new Digital Road and the Dead Reckoning, to improve the
technology. Though the paper will not go in-depth performance that of TDOA. This paper discusses on the
technically about how IPTV works, it will provide a good methods and feasibility of the TDOA-based assisted
overview of IPTV architecture, how it will be delivered from technologies application and introduces the Vehicle
the service providerâs head-end equipment to the end-user Management System concerned. It has been compared
(the consumer), and the differences between IPTV with the traditional sole positioning algorithm and given the
architecture and legacy cable TV (television) architecture. simulation. The future prospect of this system is discussed."
Most importantly, it will describe the importance of why
moving to IPTV is a competitive âhave-toâ for service 2B-11.6 1568963653
providers to survive."
Performance Improvement of Flooding Algorithm
2B-11.3 1568965123 in Highly Mobile Networks
Validation of the Random Waypoint Mobility Young Chul Park
Model Through a Real World Mobility Trace "In this paper we propose a modified flooding algitrithm for
highly mobile networks. This scheme minimizes the call set-
Andres Rojas, Philip Branch, Grenville Armitage up time and does not have to maintain routing tables.
"User mobility models are used in simulations of mobile Because of its significant drawbacks with respect to the
communications systems to study various characteristics of network efficiency, it has not found widespread applications.
network performance. One of the models which is in Using the minimum first-derivative length (MFDL) path, we
common use is the Random Waypoint Model (RWP). The improve the blocking probability performance of a circuit-
RWP is a simple mobility model based on random switched grid network without increasing the call set-up time
destinations, speeds and pause times. The RWP is often and processor loading of the algorithm."
criticised as not being representative ofhow humans
actually move. Paradoxically, validation of the RWP against
real mobility data is seen as being difficult due to the
impracticalities of obtaining real mobility data. We give
details of a real world user movement trace from which we

Tencon 2005 Page 77

2B-11.7 1568963566 2B-11.10 1568965344
SERA: Service Ratio Based Preemption for Processors Communications in the Context of
Minimizing Throughput Loss in Networks Virtual Switch Framework Using the CAN-IOP
Chun Hau Lau, Sanjay Bose, Boon Hee Soong Protocol
"Resource allocation is a fundamental problem in the Tarek GUESMI
connection oriented network environment. One key problem "A CORBA-compliant Object Request Broker (ORB) can
is to address the situation when the requested bandwidth facilitate data communication between dissimilar hardware,
exceeds the capacity of the network link. Decision has to be dissimilar operating systems, and modules written in
made to reject the new request or preempt any existing different languages. Thus, in our proposal all control and
connections in order to admit the new request. In the event management communication occurs within the context of
that preemption is triggered, multiple existing connections the ORB. Our idea is based on a conception of a free
may need to be preempted in order to free sufficient architecture as a CORBA-based middleware for distributed
resources for the new request. The time invested and work embedded systems built on the CAN bus that realize a
done for existing connections may be lost. This will affect virtual switching architecture. This platform has to connect
the network revenue as well as its overall throughput. In this different physical network types. Thus we propose a
paper, we present a decentralized preemption strategy solution to make it to behave as a router or a switcher. In
(SERA) to minimize any loss of throughput. We propose to our proposed framework we will use an ORB that may be
use the ratio of the connections remaining holding time and conform to the forthcoming Real-time CORBA standards to
its elapsed time as the suitable preemption criteria. A be embedded in a network switch. Our target switch control
connection that has just started will be better to preempt environment includes support for Virtual Switching Interface
since the work and time invested in it are minimal. Finally, (VSI) which is a protocol that allows multiple independent
both the simulation and analytical results confirm that this signalling processors to interact with a switch control
preemption strategy can significantly reduce the processor. The scope of this paper is to conceive an
degradation that would otherwise occur in the throughput." application that controls a such switch using the VSI
messaging protocol within the context of embedded ORB
2B-11.8 1568965080 over a free Distributed and Realtime Embedded (DRE)
architecture built on the CAN bus."
Link Layer Mitigation in Rural UHF-MIMO Linking
Systems 2B-11.11 1568965163
Milind Thakur, Sirisena Harsha, Ian McLoughlin Optical Switching Using Cascaded Generalised
"This paper investigates the effect of link-layer error Mach-Zehnder Switches
mitigation support in UHF-MIMO based digital linking
systems, specifically investigating the example of the Tait Laurie Cahill
Electronics rural linking system. This provides network "Optical switches based on planar lightwave technology
connectivity to remote communities using an IP-over air offer stable and robust means of achieving space-division
packet-based TDD MIMO radio system. Analytical switching. One such candidate is the generalized Mach-
expressions are derived for packetised data transfer over Zehnder optical switch. This switch normally comprises two
such a system, and simulation results presented to verify multimode interference (MMI) couplers connected by an
performance in terms of application delay experienced by array of phase shifters. However, such a switching structure
remote users under various error conditions. Results are does not possess the non-blocking characteristics required
shown that quantify the degree of improvement available in most telecommunication applications. In this paper it is
through the incorporation of link-later error mitigation shown that, if the architecture is extended to include
through selective repetition of erroneous sub-packets." additional MMI couplers, then the resultant cascaded
generalised Mach-Zehnder switch can provide improved
2B-11.9 1568965340 blocking behaviour."
Hybrid P2P Based Self Organising Grids for
Efficient Resource Distribution
Aparna Nagargadde, Srividya Gopalan, Sridhar
Varadarajan
"Resource discovery and resource distribution are
fundamental in deploying and executing applications
efficiently on a grid. Efficient resource discovery and
scheduling of applications within a grid demands additional
services like identifying the right kind of resources with
minimum delay and reduced overhead due to resource
management messages across the grid. In this paper we
employ the hierarchical overlay network for efficient
resource distribution within a grid, and adapt hybrid P2P
techniques for scheduling grid applications and self
reorganization of the grid network. These job management,
resource management, and self organisation modules are
implemented with support from existing grid middleware
and an implementation framework is presented for the
same"

Tencon 2005 Page 78

 2B-12
Session Title: Tencon Poster Session 2
Chairperson: Enn Vinnal
Day: Wednesday
Date: 23rd November 2005
Time: 10:50AM – 12:30PM
Room: M2

2B-12.1 1568964623 2B-12.3 1568965393

Camera-based 3D Tracking System TFRC Fairness Improving Using the Dynamic
Nazim Mir-Nasiri Loss Rate Measurement
"Camera-based systems are frequently used to track Yong-Whan Ho, Lee Jeong Kyoon, Ki Young Lee
moving objects which are in the field of their view. This "In the recent years, demands of real-time audio/video
paper describes design and development of a camera- stream applications have been increased. Many researches
based visual system that can constantly track moving object make progress actively for the new demands, and
without necessity of calibrating camera in real world TFRC(TCP-Friendly Rate Control) is one of the main
coordinates. This reduces the complexity of the system and stream in this development. This transport protocol which
processing time due to the unnecessary conversions and depends on TCP throughput equation, have been improved
calibrations. The system consists of a two-motor pan-tilt smoothness in transporting multimedia data and fairness
camera driving mechanism, PCI image acquisition board, with TCP. In this paper, we proposed a new loss rate
and PWM-based DC-motor driver board. It uses image calculation method that changes the weight depending on
processing techniques to identify and locate the object in the network state as a core of the TFRC protocol. In the
3D scene and motion control algorithms to direct the original TFRC method, the loss rate calculation does not
camera towards the object. Thus the objective of the project consider the network state and the weight is unchanged. As
is to develop a vision system and control algorithms which a result, in a heavy-congested network, the fairness of
can be locked on the moving object within the field of its TFRC falls off rapidly and it makes network state worse.
view. The developed software and related interface The proposed method gives more weight to the latest loss
hardware monitor and control the motors in such a way that interval, when the loss interval has decreased for 5 intervals
the moving object should always be located right at the or more in a row. In the original method, the latest loss
center of the camera image plane. The system, in general, interval is only assigned 16% of the total weight, regardless
simulates 3D motion of the human eye when it always tends of the current congestion level. The proposed method gives
to focus on the moving object within the range of its view. It more weight to the recent loss interval during the congested
actually imitates the tracking ability of human eye." period. When the proposed method is used, the latest loss
interval is assigned 61% of the total weight in calculating
2B-12.2 1568965085 the loss rate. In a congested network, TCP's normalized
throughput decreases less than 0.8 with the original
Assembly Modelling Through Constraint-based method, but the proposed method can maintain the
Manipulations in A Virtual Reality Environment throughput more than 0.8 and enables the TCP to exist
Yongmin Zhong, Bijan Shirinzadeh, Weiyin Ma, along with the TFRC."
Hwee Choo Liaw
2B-12.4 1568964923
"With todays Virtual Reality (VR) systems, it is difficult to
create precise assembly models through 3D interactions On PAPR Reduction in OFDM Systems: A
due to hardware limitations. In this paper, an efficient Technique using Normalized Complex Hadamard
approach is presented for intuitive and precise assembly Transform
modeling in a VR environment. Assembly modelling is
performed by precise constraint-based 3D direct Kuan Min Yew, Varun Jeoti, Micheal Drieberg
manipulations in an intuitive manner. Constraint-based A new technique for the reduction of peak to average power
manipulations are accompanied with automatic constraint ratio (PAPR) in Orthogonal Frequency Division Multiplexing
recognition and precise constraint satisfaction, and are (OFDM) systems is presented. It is based on precoding the
realized by allowable motions for precise 3D interactions in constellation symbols with Normalized Complex Hadamard
the VR environment. Some special constraint-based Transform (NCHT). The proposed technique can achieve
assembly operations are constructed for assembly about 3 dB more reduction in PAPR compared to
modelling in the VR environment. A method for recognizing conventional OFDM systems. The technique has been
the pairs of mating features between assembly components shown to be effective and has low complexity. The NCHT
is presented. This method integrates 3D direct can be a promising technique for PAPR reduction of OFDM
manipulations with a feature mating knowledge base and systems.
makes the process of recognizing a pair of mating features
intuitive. A concept of offset solid is also presented for 2B-12.5 1568965137
determining assembly components. A prototype system has
been developed for assembly modelling through precise Blind Channel and Carrier Frequency Offset
constraint-based manipulations in an intuitive manner in the Estimation for Zero-Prefix OFDM Systems
VR environment."
Qi Cheng
"In this paper, a blind algorithm is proposed for channel and
CFO estimation in zero-prefix OFDM systems. By exploiting
the greatest common divisor structure of the received signal
for two or more symbols, the algorithm yields joint estimates
of channel parameters and carrier frequency offset. It
provides error-free message recovery at moderate SNRs."

Tencon 2005 Page 79

2B-12.6 1568965142 2B-12.9 1568965034
Performance of OFDM Carrier Frequency Offset Development of Power System Analysis Software
Estimation Based on Multiple Blocks Using Object Components
Qi Cheng Khalid Mohamed Nor, Hazlie Mokhlis, Hadi Suyono,
"This paper examines the performance of a method for Mamdouh Abdel-Akher, Taufiq Ghani, Abdul Halim
carrier frequency offset estimation. Although the method is Abdul Rashid
derived from a maximum likelihood framework, its "This paper presents experiences of developing a power
performance no longer achieves the Cramer-Rao bound system analysis software using a combination of Object
when multiple OFDM blocks are in use. The phenomenon is Oriented Programming (OOP) and Component Based
verified using statistical estimation theory. An improved Development (CBD) methodologies. In this development,
method is also proposed." various power system analyses are developed into software
components. These components are integrated with
2B-12.7 1568965065 graphical user interface components to build up a power
Automatic Offset Calibration of Quadrature system analysis application. By using both OOP and CBD
methodologies, updating or adding new algorithm can be
Encoder Pulse Sensor For Vector Controlled done to any specific component without affecting other
Drive of Permanent Magnet Synchronous Motors components inside the software. The component also can
Mongkol Konghirun be replaced with any other better component whenever
necessary. Hence, the software can be maintained and
"Quadrature Encoder Pulse (QEP) sensor has been widely updated continuously with minimum resources. The
used to detect the rotor position of Permanent Magnet performance of the components is described in comparison
Synchronous Motors (PMSMs) in the vector controlled drive with the non-component applications in terms of reuse as
systems. However, the information regarding to the angular well as execution time."
difference between the index signal and the back emfs of 2B-12.10 1568965176
motor (so called the index offset) is sometimes lacked in the
manufacturers datasheets. As a result, by using the Applications of FACTS Controller for Improving
incorrect offset, the back emfs are not in phase with the Power Transmission Capability
stator currents. Then, the produced electromagnetic torque
is not optimal in the vector controlled drive system of PMSM Elena Yap, Majid Al-Dabbagh, Peng Chew Thum
using QEP sensor. This paper presents a practical and "This paper focuses on the effective utilization of a Flexible
simple technique to automatically calibrate this offset. The Alternating Current Transmission System (FACTS) device
experimental results will also demonstrate the effects of called Unified Power Flow Controller (UPFC) for power flow
incorrect offset to the torque production. The overall drive control. Its main components, the Static Shunt
system is implemented on a 32-bit fixed-point digital signal Compensator (STATCOM) and the Static Series
processor based controller (TMS320F2808eZdsp board)." Synchronous Compensator (SSSC) are actually shunt and
series compensators respectively. This study demonstrates
2B-12.8 1568965450 the use of the latest Power System Analysis Toolbox
(PSAT) package for network analysis of alternative means
Co-optimizing Ancillary Raise and Lower of improving existing transmission capability. The IEEE 14-
Services bus network system operating with common loads and
Pramesh Chand generators is used to demonstrate the available
controllability in the UPFC through computational
"This paper describes an innovative approach of co- monitoring. An analysis between FACTS and High Voltage
optimizing the energy and ancillary raise and lower services Direct Current (HVDC) are also presented in this paper for
markets in the Australian National Electricity Market. A comparison in power quality and cost-efficiency. The case
heuristic based technique called Horizon Scan [1], [2] is study provides a quick steady-state power flow state
applied as the optimization procedure to find optimal condition solution using the PSAT."
solution for the problem. Horizon Scan is a heuristic based
technique to search for optimal solution in a non-linear 2B-12.11 1568965478
solution space. The paper is an extension to work that has
been originally presented [3] and [7]. It includes a case A New Approach for Generation Maintenance
study depicting simplified two region market model. Optimal Scheduling in a Deregulated Power System
schedules resulting in economical dispatch costs are
obtained when Horizon Scan is applied to optimize trading Nima Mohammadi Tabari
between multi-regions with inter-connectors. Linear and "This paper presents a dynamic programming methodology
non- linear constraint equations are also included in the for finding the optimum preventive maintenance schedule of
model to take into account dynamics of the power system." generating units in a deregulated power system. The
objective function is to maintain the units as earlier as
possible. Various constraints such as spinning reserve,
duration of maintenance and maintenance crew are being
taken into account. In a case study, using the power system
of [1] with six generating units, and implying dynamic
programming, we obtain one optimum solution and seven
other feasible solutions."

Tencon 2005 Page 80

 2C-01
Session Title: Network Performance and Design 2
Chairperson: Alexander Kist
Day: Wednesday
Date: 23rd November 2005
Time: 2:00PM – 3:40PM
Room: M9

2C-01.1 1568965140 2C-01.3 1568965252

An optimum mapping of IPs for On-Chip Network Optimized Pricing and Closed Form Algorithm for
design based on the minimum latency constraint WFQ Scheduling
Ngo Duc, Hae-Wook Choi Lari Kannisto, Ari Viinikainen, Jyrki Joutsensalo,
"A new chip design paradigm, so called Network on Chip, Timo Hämäläinen
has been introducing based on the demand of integration of The quality of future network services can be guaranteed
many heterogeneous semiconductor intellectual property with appropriate pricing and packet scheduling schemes. In
(IP) blocks. The Network on Chip design not only requires this paper a closed form formula for updating adaptive
the good performance but also the minimization of several weights of a packet scheduler is derived from a revenue-
physical constraints such that the network latency, the used based optimization problem. The algorithm ensures
area as well as the power consumption of design. This bandwidth as a Quality of Service (QoS) requirement and
paper analyzes the average latency of heterogeneous optimizes revenue of the network service provider. We
Network on Chip architectures in which the shortest path derive and prove the guaranteed bandwith and revenue
routing algorithm is applied. This average latency includes optimization algorithm which has been tested with NS-2
the queuing latency and the wire latency, and is calculated simulations. The weight updating algorithm is implemented
for general cases of allocating IPs onto a fixed generic with a network simulator to compute the weights of a WFQ
switching architecture as 2-D Mesh. With different allocation scheduler and compared to constant weights.
schemes of IPs, the network has different average
latencies. Hence, this article presents an optimal search 2C-01.4 1568965332
that adopts the Brand and Bound algorithm to find out the
optimal mapping scheme to achieve the minimal network Analysis of performance improvement with
latency. This algorithm automatically map the desired IPs packet dispersion
onto the target Network on Chip architecture with the
Fumio Ishizaki
criteria of lowest network latency. The optimal latency, as
our research target, allows us to achieve the design with "This paper analytically studies performance improvement
low power consumption as well. We also carry out the with packet dispersion. Contrary to the previous works on
experiment for On Chip Multiprocessor Network application. packet dispersion, we consider the tail distribution of queue
The results shows that 36.5% and 30.8% of network latency length and that of packet delay as performance measures,
is saved with the optimized allocation scheme for the cases and we model a packet network as multiple parallel queues
of 3x3 and 4x4 Mesh architectures of On Chip where the arrival processes from sources are not renewal
Multiprocessor Network application, respectively." but highly bursty. In the numerical results, we observe that
packet dispersion can greatly improve the delay
2C-01.2 1568965198 performance of packets. We also see that packet-level load
balancing is superior to flow-level load balancing."
Block Based Window Retransmission ARQ
Scheme for 100Mbit/s Infrared Links 2C-01.5 1568965504
Mohammad Shah Alam, Mst. Shamim Ara Shawkat, Connectionism-Based CBR Method for
Mitsuji Matsumoto Distribution Short-Term Nodal Load Forecasting
"With the increase of data transfer rates beyond 100Mbit/s Jianzhong Wu, Yu Yixin
over the infrared links, a robust automatic repeat request
(ARQ) scheme is necessary to cope with the high speed "Short-term nodal load forecasting is very important for
environment. At high data rate, the Go-Back-N (GBN) ARQ operation, planning and management of distribution
scheme requires window and frame size adaptation to the networks, which is characterized by lack and imprecision of
corresponding optimum values for the correspondent Bit historical data, inconspicuous trend for load variation, and
Error Rate (BER). But adaptive approaches always add a changeful pattern of nodal load. It is very hard for
significant amount of complexity to the system. This paper conventional methods to solve such problem. A
presents a non adaptive Block Based Window connectionism-based case-based-reasoning method
Retransmission (BBWR) ARQ scheme that operates with a (CBCBR) is proposed based on Parallel Distributed
fixed size receiver buffer and achieves a significant Processing (PDP) model. The principle of CBCBR is
improvement in the throughput efficiency over a wide range analyzed, the elementary architecture of CBCBR is defined,
of BER at 100Mbit/s infrared links. Simulation results are and a hybrid supervised/unsupervised learning algorithm,
presented which also reveal the effect of link layer which equips CBCBR with a good generalization capability,
parameters, such as window and frame data length, and is also proposed. CBCBR can build nodes and connections
physical layer parameters, such as minimum turn around dynamically by the rapid and incremental learning
time on the performance of the proposed scheme." procedure and can withstand the effect of bad data
effectively through network self-organizing. The proposed
method is tested using load data of a practical system and
the test results is compared with that of BP network, RBF
network and AR model."

Tencon 2005 Page 81

 2C-02
Session Title: Multimedia 1
Chairperson: Leith Campbell
Day: Wednesday
Date: 23rd November 2005
Time: 2:00PM – 3:40PM
Room: M7

2C-02.1 1568963317 2C-02.4 1568965044

Ontology-Based Multimedia Presentation Multisource Receiver-driven Layered Multicast
Generation Liang Zhao, Hideo Yamamoto
Payam Mamaani Barnaghi, Sameem Abdul Kareem "Receiver-driven Layered Multicast (RLM) is an efficient
"Multimedia data are illusory entities for the machines. Their solution for streaming multimedia content over the IP
contents include interpretable data as well as binary network. In RLM, a content is encoded into a number of
representations. Understanding and accessing the content- layers that are sent to distinct multicast groups
driven information for multimedia objects allow us to design simultaneously. On the other hand, a receiver subscribes to
an efficient multimedia querying and retrieval system. In this a subset of groups according to some criterion (e.g.,
paper, we propose a framework to represent the multimedia network congestion status) with no direct interactivity with
information and object roles in order to generate automatic the sender. This paper considers Multisource RLM (MRLM)
multimedia presentations. The proposed architecture in which multiple content sources are available and a
attempts to represent the semantic information and the receiver has to subscribe in an RLM manner but has an
relations amongst the multimedia objects in a disclosure additional criterion -- content source priorities. This problem
domain. Thus, the system is domain dependent. The has applications in remote monitoring and many other
represented data associates with the presentation fields. We propose a reception control method called PAC
mechanisms to create an integrated presentation (Priority-based Autonomous Control).The main idea of PAC
generation system. A multi-layer design defines the various is to allocate active time periods for receiver processes
levels of abstraction for the proposed framework." according to given priorities while leaving each process run
autonomously. A preliminary experimental study is shown to
2C-02.2 1568963327 demonstrate algorithm PAC."

A Novel Grid-based Content Management System 2C-02.5 1568965055

Qian Zhang, Yu Sun, Xuezhi Wen Kansei-based investigation of methods for
"Content Management Systems (CMS) determine a projection of digital content
valuable means to simplify the creation, maintenance, and
distribution of content. However, the state of the art of CMS
Toshifumi Mori, Tomohiro SHIMIZU, Shota
is still at its infant stage and confronts with many challenges YAMANE, Mie Sato, Masao KASUGA
such as security, replica consistency, reliability and "More and more video users now enjoy the sense of a
scalability. To address the challenges, this paper proposes gspecial seath given by modern video devices thanks to the
a novel Grid_based content management system (GCMS) development of broadcast technology and digital
with some Peer-to-Peer (P2P) properties such as technology. There are also increasing opportunities for
scalability, robust, etc.,which combines the advantages of users to view video presentations in the work environment.
client-server model and pure P2P model and avoids their When different types of video equipment are used,
shortcomings. The paper also introduces an approach for however, their technical characteristics affect how the
providing the strong and efficient access control, which is a content is displayed; it is not uncommon for viewers of the
key requirement for CMS." presentations under different media to be left with differing
impressions. This research is intended to point out
2C-02.3 1568965003 directions for development of systems which will provide
viewers with the same sensory impressions of any given
Quantified Cost-Balanced Overlay Multicast video production, regardless of the mode of viewing. The
Routing for Multimedia Network present report focuses on 2 projectors of differing types.
Jun Lu, Rongrong Ni The projectors were adjusted to minimize differences in
color reproduction; the investigation examined sensory
"This paper focuses on the quantitative analysis issue to the assessments of the perceived differences between the
overlay multicast routing metrics tradeoff problem, and display modalities."
presents a quantified cost-balanced overlay multicast
routing scheme (QCost-Balanced) to the metric tradeoff
problem between overlay path delay and access bandwidth
at Multicast Server Nodes (MSN) for real-time applications
over Internet. Besides implementing a dynamic priority to
MSNs by weighing the size of its service clients for better
efficiency, QCost-Balanced tradeoffs these two metrics by
using a unified tradeoff metric based on quantitative
analysis. Simulation experiments demonstrate the scheme
achieves a better tradeoff gain in both two metrics, and
effective performance in dynamic metric quantitative
control."

Tencon 2005 Page 82

 2C-03
Session Title: Wireless LANS 1
Chairperson: Ahmet Sekercioglu
Day: Wednesday
Date: 23rd November 2005
Time: 2:00PM – 3:40PM
Room: P2

2C-03.1 1568963545 2C-03.4 1568965170

The 802.11 Technology Gap Case Studies in Enhancing Data Efficiency in IEEE- 802.11a Using
Crime Frequency and Time domain Equalization
Benjamin Turnbull, Jill Slay Vijay Kumar Chakka
"Much has been written discussing the security "In this paper, increase in data rate is achieved by removing
vulnerabilities with 802.11 based networks, and what the cyclic prefix between the OFDM symbols. The
constitutes legal and illegal activity within the area. intersymbol interference and intercarrier interference
However, there is little academic literature available caused by the removal of cyclic prefix are removed by
discussing whether this technology is being exploited for equalization. We propose two methods of equalization. In
criminal purposes, and if it is, for what purposes. This paper the first method, the zero frequency domain equalization
investigates all public prosecutions where the accused have (ZF-FEQ) is done by making use of spill over in the unused
misused 802.11 wireless networks and the consequent carriers. The pre-FFT time domain Equalizer minimizes the
outcomes. From this, it can be observed how similar crimes noise amplification caused by ZF-FEQ. In the second
are being investigated, and possible future crimes may be method, the Inter Symbol Interference and Inter Carrier
examined." Interference are removed at the transmitter itself by
transmitting data in the unused carriers, which is the
2C-03.2 1568963626 combination of data in the used carriers. The requirement of
power in the unused carriers increases as the length of the
Performance of IEEE 802.11b Wireless Links: An channel increases. Using the TEQ, the effective length of
Experimental Study the channel is reduced to minimum. The simulation results
Duc Pham, Ahmet Sekercioglu, Gregory Egan are presented in support of these two methods."

"This paper reports the experimental results of IEEE 2C-03.5 1568965240

802.11b wireless link performance in ad hoc mode. The
throughput and packet loss rate are measured for User Path-Based Minimum Power Broadcast Problem
Datagram Protocol (UDP) and Transmission Control in Wireless Networks
Protocol (TCP) traffic in both indoor and outdoor
environments. Various factors that affect the performance of
Yean-Fu Wen
the wireless link, such as the packet size, the transmission "Broadcasting is the most common method of transmission
rate, the distance between stations, the retry limit and the for shared media in local area networks. The radius range
network topology are considered. Experimental determines the number of hearing nodes that can be used
measurements are also compared against theoretical to deliver multicast messages and replace multiple copies
calculations." with the same packet through unicast in wireless networks.
The crucial design challenge is energy efficiency, because
2C-03.3 1568964072 each individual node only has a small battery as a power
source. Thus, the objective of this paper is to find the
An optimized DCF schema with Downlink and optimal radius range so that the total power consumption is
Uplink bandwidth control in 802.11 WLAN minimized. The problem is formulated as a minimum power
Hui Feng, Ting Wang, Bo Hu broadcast tree constructed based on paths instead of links
or nodes. Since this problem is NP-complete, we adopt
"In 802.11 infrastructure BSS, the channel throughput will Lagrangian Relaxation (LR) to decompose it and solve the
decrease when increasing the number of competing STAs. sub-problems. The LR dual-mode problem ensures the
Also severe asymmetry exists between downlink and uplink objective lower bound value. The primal mode problem is
data because AP accesses the share channel without any solved via our proposed approximation heuristic, which
priority. Both problems are solved by a new MAC schema takes prompts from a set of LR multipliers, to obtain the
which adjusts the initial contention window of AP and STA upper bounds objective value. We present experimental
to optimized values. A new estimation method for STA results from randomly generated networks and show that
number is also proposed via dynamic MAC addresses our proposed algorithm saves more than 30%, 5%, and
statistic at AP side. Afterwards, optimized initial contention 10% energy compared to the Prims minimum spanning tree
window values are derived from a mathematical model (PMST), the Broadcast Incremental Power (BIP), and
under downlink and uplink bandwidth control. Simulation another proposed Greedy Incremental Broadcast Tree
indicates that the schema proposed can estimate the (GIBT) algorithms, respectively.n"
channel situation exactly and improve channel throughput
with downlink/uplink bandwidth control obviously."

Tencon 2005 Page 83

 2C-04
Session Title: Image Processing 4
Chairperson: Henry Wu
Day: Wednesday
Date: 23rd November 2005
Time: 2:00PM – 3:40PM
Room: P1

2C-04.1 1568965334 topological information, the k-nearest clustering method is

adopted to find a local neighborhood for each sample point,
Development of Intelligent Wheelchair System and a surface normal at each sample point is estimated by
based on Stereo Omni-directional System (SOS) fitting a tangent plane in its neighborhood. We construct
Yutaka Satoh, Caihua Wang, Katsuhiko Sakaue adaptive multiresolution hierarchies for several point-
sampled models and display them based on the point-
"This paper presents a novel vision-based intelligent based rendering. Experimental results show this approach
wheelchair system for supporting independent movement of is efficient for rendering complex point-based 3D models."
physically handicapped people. A key device of the system
is Stereo Omni-directional System (SOS) which can capture 2C-04.4 1568965462
all-directional color and stereo images in real time with a full
field of view of 360x360 degrees. By applying the camera Enhancement of Digital Images Using Stochastic
system to an electric wheelchair, we are aiming at Resonance
development of functions of autonomous obstacle-
avoidance control and gremotely accompanyingh a
Rajib Jha, Prabir Kumar Biswas, B. N. Chatterji
handicapped person. The present paper focuses on "In certain non-linear dynamical systems, the addition of
fundamental technologies of important note, such as auto- noise can assist the detection of a signal. In general, noise
correction of camera orientation and high-speed, high-grade is always seen as deteriorating the system performance
integration of images from all directions, from the viewpoint very effectively and is therefore normally considered as a
of mounting SOS on an electric wheel chair." nuisance. However, in certain non-linear systems including
non-linear signal and image processing, the presence of
2C-04.2 1568965356 noise can in fact enhance the weak signal strength or the
low contrast images. This phenomenon is termed stochastic
Rapid 3D Simultaneous Algebraic Reconstruction resonance. In this paper, we present a novel approach for
Technique (SART) on Clustering System contrast enhancement of low contrast image, which
KOSIN KALARATT outperforms the other image enhancement techniques.
Block diagram of our algorithm for stochastic resonance
"An important problem in image processing is to construct a and derivation of threshold equations to be used in image
cross section of an object from several images of its trans- contrast enhancement is also presented in this paper. The
axial projection. However, the time consuming and the result of this processing yields a contrast-enhanced image
complexity of the reconstruction process are the crucial with better visualizibility, no blocking artifacts and no spot in
problems. In addition, the reconstruction process requires an image. Results of our methods are compared with the
very high performance of the computer. Therefore, in this enhancement method based on histogram equalization and
paper, concepts of parallel programming method is wavelet transform."
employed to speed up a 3D Simultaneous Algebraic
Reconstruction Technique. The scheduling process in the 2C-04.5 1568965491
clustering system is improved. The appropriate amount of
work is distributed to each computer (node) in the clustering Restoration of Non-Uniformly Warped Images
system using a work-pool scheduling scheme. In this Using a Typical Frame as Prototype
scheme, even though the work load for each node in
clustering system is uneven, the variation of the processing
Murat Tahtali, Donald Fraser, Andrew Lambert
time for each node is insignificant. Our proposed system "When imaging through the atmosphere, the resulting
works successfully with decreasing the reconstruction time image also contains a combination of non-uniform blurring
up to 75% percent referred to the normal image and random shifting of each point. We are proposing in this
reconstruction form projection performed on a single paper an alternative and potentially more accurate method
computer." of restoring the geometry, which is better suited to real-time.
In the absence of a truth image to estimate the warping, the
2C-04.3 1568965436 established method of obtaining this prototype is by
averaging a large number of individual frames. Geometric
Adaptive Multiresolution Representation of a accuracy is achieved at the expense of spatial resolution.
Point-Based 3D Model for Rendering As an alternative, we are proposing to use one of the
Jihong ZHOU, Qiuqi Ruan frames. Once the warping functions in terms of shifts are
obtained, subtracting their respective average from the
"A multiresolution representation of a 3D model is one of individual shifts will give the shifts required to restore each
important means for its fast rendering. In this paper we frame. In this process, there is no degradation in the spatial
present an approach to build up an adaptive multiresolution resolution of the prototype; therefore a more accurate
representation for a point-sampled 3D model directly from estimation of the warping functions can be expected."
its set of sample points. The binary space partition (BSP) is
used to recursively split a point¨Csampled model according
to the covariance analysis of its sample points. Thus a
model is subdivided along the direction of the greatest
variation, and the subdivision is adaptive to the spatial
distribution of sample points. A multiresolution hierarchy
based on bounding spheres is constructed for a model from
its subdivision. Since a point-sampled model has not any

Tencon 2005 Page 84

 2C-05
Session Title: IT Applications 4
Chairperson: Paul Beckett
Day: Wednesday
Date: 23rd November 2005
Time: 2:00PM – 3:40PM
Room: P3

2C-05.1 1568965400 2C-05.4 1568965530

Software Reliability Modeling with Weibull-type A Bus Level SystemC Model for Evaluation of
Testing-Effort and Multiple Change-Points Avionics Mission System Data Bus
Chu-Ti Lin, Chin-Yu Huang Rajavadivel Chinnapichappan, Syed Aziz
"Software reliability is defined as the probability of failure- "This paper presents a bus level SystemC model of an
free software operation for a specified period of time in a avionics mission system data bus. The mission system is a
specified environment. Over the past 30 years, many very complex distributed computer network consisting of
software reliability growth models (SRGMs) have been mission control computer, radars, an array of subsystems
proposed for estimation of reliability growth of products and sensors. The data bus plays a central role in the
during software development processes. SRGMs proposed system as it carries all the information between the system
in the literature took into consideration the amount of components. Therefore modeling the bus at an appropriate
testing-effort spent on software testing which can be level of abstraction using appropriate technology is very
depicted as a Weibull-type curve. However, in reality, the important for evaluation of the performance of both the bus
consumption rate of testing-effort expenditures may not be and the entire system. We explore the benefits of using
a constant and could be changed at some time points. SystemC in modeling such a critical data bus. The
Therefore, in this paper, we will incorporate the concept of developed model has been validated for various payloads
multiple change-points into the Weibull-type testing-effort and system size. The enhanced features of SystemC
function. New model is proposed and the applicability of makes the model more efficient in some respect compared
proposed model is demonstrated through two real data to models developed using conventional hardware
sets. The experimental results show that this model has a description languages."
fairly accurate prediction capability."
2C-05.5 1568965556
2C-05.2 1568965426
A Nanowire Array for Reconfigurable Computing
Implementation and Evaluation of the Extended
Paul Beckett
Cell Splitting Algorithm
"A fine-grained reconfigurable array based on
Michiko Tanaka, Kunihiko Kaneko, Yingliang Lu, complementary, dual-gate, fully depleted, silicon on
Akifumi Makinouchi insulator (DGFDSOI) nanowire transistors is proposed and
"We herein present the extended cell splitting algorithm, analyzed. Both low power and reconfigurable operation may
which first splits bounded or unbounded one-dimensional be achieved by altering the switching threshold of the array
cells, one at a time, and splits cells of dimensions greater using the back-gate bias on the complementary double-gate
than one. An evaluation of the algorithm is also presented." transistors. Simulated performance figures are presented
for the array when configured into representative circuits
2C-05.3 1568965437 and compared with two similar self assembled molecular
arrays. It is shown that SOI nanowire arrays can achieve
M-ary Chirp Spread Spectrum Modulation for dense, low-power reconfigurable operation without the
Underwater Acoustic Communication overheads of either level restoration or additional gain
blocks that may be required by molecular-based systems."
Jianguo Huang, He Chengbing, Zhang Qunfei
"A M-ary chirp spread spectrum modulation (MCSS) for
medium range underwater acoustic (UWA) communication
is proposed in this paper. The method of MCSS uses M-ary
different chirp signals to represent message at the
transmitter and a bank of correlators to demodulate the
message at the receiver. There are three advantages for
MCSS applied to UWA communication: good performance
against multi-path interference and channel fading, higher
data transmission rate with robustness and larger
processing gain. Experimental results in the lake show the
efficiency for MCSS that is in the distance of 25km, the data
rate can reach 300 bps with BER below 10-4."

Tencon 2005 Page 85

 2C-06
Session Title: Power Quality Monitoring 2
Chairperson: Chengshan Wang
Day: Wednesday
Date: 23rd November 2005
Time: 2:00PM – 3:40PM
Room: M3

2C-06.1 1568964892 2C-06.4 1568965151

Intelligent Classification of Power Quality Maximising Voltage Support in Distribution
Disturbances in a Distribution Network Systems by Distributed Generation
Bijaya Panigrahi, Sukumar Mishra An Le, Mohammad Kashem, Michael Negnevitsky,
"This paper presents a new approach for power quality Gerard Ledwich
analysis using a phase corrected wavelet transform "Rapidly increasing in the demand of electricity along with
(PCWT) known as S - Transform. The local spectral recently growing advancements in Distributed Generation
information of the wavelet transform can, with slight (DG) technologies have sparked a new interest in utilisation
modification, be used to perform local cross-spectral of DG sources. However, there is a number of technical
analysis with very good time resolution. The phase issues regarding to the installation of DG, since DG
correction absolutely references the phase of the wavelet connection significantly affects existing network
transform to the zero time point, thus assuring that the configuration and operation, especially for system with high
amplitude peaks are regions of stationary phase. The level of DG penetration. To achieve maximum voltage
excellent time frequency resolution characteristic of the support from DG utilisation, methodologies related to DG
PCWT makes it an attractive candidate for analysis of placement and DG operation are developed in this paper.
power system disturbance signals. Several power quality Technique to optimise voltage improvement by effectively
problems are analyzed using PCWT and it clearly injecting active and reactive power of DG is developed
demonstrates the advantages of PCWT in detecting, based on voltage sensitivity of lines. An index has been
localizing and classifying the power quality problems. The developed to obtain optimal or near optimal placement of
features extracted by using PCWT are trained by a radial DG for maximum voltage improvement in a distribution
basis functional link net to classify the PQ events" feeder. Simulation studies are conducted on IEEE 16-bus,
33-bus and 69-bus radial systems to verify the developed
2C-06.2 1568964999 method and results are reported. Index Terms-- Distributed
Generation (DG), Sensitivity, Power Distribution, Optimal
Power Quality Disturbance Recognition Using Placement, DG Operation, Voltage Support."
Wavelet-Based Neural Networks
Suriya Kaewarsa, Kitti Attakitmongcol 2C-06.5 1568965342
"This paper proposes a wavelet-based neural network Development of Efficient Algorithm for Fault
classifier for recognizing power quality disturbances is Location on Series Compensated Parallel
implemented and tested under various transient events. Transmission Lines
The discrete wavelet transform technique is integrated with
multiple neural networks using a learning vector Sarath Kapuduwage
quantization network as a powerful classifier. Various "Development of efficient algorithm on series compensated
transient events are tested, the results show that the parallel transmission line is presented. The new algorithm is
classier can detect and classify different power quality developed using instantaneous fault data collected from
disturbance types efficiency." both ends of the line, in order to estimate the fault location
and fault resistance accurately. Firstly, single transmission
2C-06.3 1568960546 line with series compensation unit, which is located in the
Compression of Power Quality Disturbance Data middle, is modelled using ATP program and tested with the
new algorithm with number of different fault cases. Then the
Based on Energy Threshold And Adaptive algorithm further improved for the testing of parallel series
Arithmetic Encoding compensated parallel transmission lines. A sample example
Jidong Wang, Chengshan Wang of estimation of fault location and resistance on single line
model is reported and discussed in this paper, and a
"Recently, power quality issues have captured more comprehensive summary of fault location estimations for
attention. It is necessary to monitor power quality in order to single line model is presented in this paper. At present, the
analyze and evaluate it. The monitors will record huge data algorithm is being tested for parallel transmission lines and
during disturbance. It is inconvenient for data storage and results obtained from trial run are satisfactory. However,
transmission. Compression of power quality disturbance further testing of parallel transmission lines case will be
data will save storage space efficiently and accelerate presented in publication."
transmission speed. This paper proposes energy threshold
method based on wavelet transform, and then integrates
adaptive arithmetic encoding to compress disturbance data.
The compression ratio is improved and performance is
better. Four typical power quality disturbances including
voltage sag, swell, interruption and transient impulse are
used to test the proposed method, the validity is verified by
simulation results."

Tencon 2005 Page 86

 2C-07
Session Title: Adaptive Systems and Signal Processing 2
Chairperson: Yunhan Dong
Day: Wednesday
Date: 23rd November 2005
Time: 2:00PM – 3:40PM
Room: M6

2C-07.1 1568964959 2C-07.4 1568965298

LMS Algorithm for Blind Adaptive Nonlinear Computer Aided-Program for Validation of
Compensation Measuring System from Unit Step Response by
Kutluyil Dogancay Time Convolution Method
"This paper presents low-complexity blind adaptive P. YUTTHAGOWITH
nonlinear compensation algorithms for bandlimited signals. "This paper presents the design and development of the
The new algorithms utilize highpass filtering to extract the computer program based on LabVIEW program [1] for
out-of-band signal energy caused by nonlinear distortion. A verification of HV measuring system. This program consists
least-mean-square (LMS) algorithm and its normalized of two parts. The first part is a program for calculation the
version are derived based on minimization of the square of parameters of unit step response which are verified ability
the extracted out-of-band signal without access to the of measuring system by comparison with the values given
original input signal or prior knowledge of the nonlinearity. in IEC 60060-2 [2]. The second part is a program for
In this sense the developed algorithms are ""blind"" and calculation the output voltage waveform by time convolution
only require prior knowledge of the signal bandwidth. Unlike between standard testing waveform and unit step response.
the $P$th-order power series inverse, the proposed The standard waveforms under investigation are full
nonlinear compensation method is not affected adversely lightning impulse voltage waveforms, tail-chopped lightning
by large input amplitudes. The effectiveness of the online impulse voltage waveforms, front-chopped lightning impulse
algorithms is illustrated with several simulation examples." voltage waveforms and steep-front voltage waveforms
according to IEC standards 60060-2 and 61211[3]. This
2C-07.2 1568965161 paper considers two different types of the measuring
systems, using the resistive voltage divider or damped
Compressed Space-Time Adaptive Processing capacitive voltage divider for testing this program. The
(CSTAP) results show that this program is practicable, and should be
Yunhan Dong a useful program for impulse voltage measurements in a
standard high voltage laboratory."
"A wavelet transform technique is used to reduce
(compress) the dimensionality of the covariance matrix as
well as signal snapshots in space-time adaptive processing
(STAP) to cut the computational demand. The algorithm is
tested using both simulated airborne radar data generated
by the high fidelity airborne radar simulation software,
RLSTAP, as well as real airborne radar data collected by
the MCARM system. It shows that while the computational
demand is reduced by as much as 85%, there is no
sacrifice to signal detection. Unlike existing dimensionality-
reduced algorithms which are based on certain
assumptions and lead to partially adaptive STAP, the
proposed algorithm, compressed STAP (CSTAP), does not
require any assumptions, so is still fully adaptive."

2C-07.3 1568965294
Adaptive Surface Fitting for Inspection of FPD
Devices Using Multilevel B-spline Approximation
Gyu-Bong Lee, Yun Su Chung, WooSeob Kim, Joon-
Jae Lee, Kil-Houm Park
"This paper proposes a surface fitting algorithm for
inspection of flat panel display(FPD) based on multilevel B-
spline approximation(MBA). FPD devices have non-uniform
background, and shape and intensity of defects are very
various. MBA is useful algorithm for surface fitting but it is
sensitive to number of control lattices. As level increases,
details of original image can be reconstructed. To detect
defect, defective area must not be well-reconstructed. In
this paper, we propose a level decision method of MBA for
FPD. After process of each level, we decide whether stop to
increase level of MBA or not with simple distance measure.
Experimental result shows that our proposed method
makes to detect defects easy."

Tencon 2005 Page 87

 2C-08
Session Title: Signal Processing Hardware 2
Chairperson: Qassim Nasir
Day: Wednesday
Date: 23rd November 2005
Time: 2:00PM – 3:40PM
Room: M4

2C-08.1 1568965301 2C-08.4 1568964936

An Asynchronous BI-Polar Ternary ADC A Novel Maximum Likelihood Frequency
Pravin Dakhole, Abhijit Bapat, Ashok Dighe Synchronizer for OFDM Systems
"A ternary ADC using multi-valued logic levels of 0, 1 and 2 Jie Zhang, Guisheng Liao, Hongqing Liu
is developed. Properties of uni-polar ternary windows are Feedback carrier frequency recovery for orthogonal
modified to suit asynchronous conversion of bipolar signals. frequency-division multiplexing (OFDM) signals transmitted
Two algorithms are presented, one giving a straight ternary over frequency selective fading channels is presented in
code uses identical digit cell topologies while the other this paper. Its derivation is based on minimum variance
gives an additional offset ternary code. P-Spice simulation conditioned maximum likelihood (MV-CML) principle. The
results of 3-digit circuits are presented." channel power on subcarriers required to perform frequency
synchronization is regarded as unknown slow time-variant
2C-08.2 1568965472 parameter and is estimated online. The performance of the
proposed synchronizer is assessed by computer
A New Convolutional Formulation of the DFT and simulations with comparison to that provided by other
Efficient Systolic Implementation schemes and Modified Cramer-Rao lower bound. Both
Pramod Meher theoretical analysis and simulation results show that our
synchronization scheme is a good candidate for OFDM
A reduced-complexity circular-convolution-like formulation is frequency synchronization.
presented for computation of the discrete Fourier transform.
A systolic architecture is also derived for VLSI 2C-08.5 1568965104
implementation of the proposed algorithm. The proposed
architecture is fully-pipelined and contains regular and Carrier Recovery Algorithm for Bandwidth-
simple processing elements. It is devoid of complex control Efficient Coded Modulations at Low Signal-to-
structure and scalable for higher transform length. It is Noise Ratios
observed that the proposed systolic structure offers nearly 4
times more throughput and significantly low latency with the Mohamed Nezami
same order of hardware complexity compared with the "A new carrier recovery algorithm for bandwidth efficient
existing structure for high-throughput DFT. modulations at low signal-to-noise ratios (SNR) is proposed
using only 5 to 10 percent of overhead symbols. It obtains
2C-08.3 1568964696 carrier offsets without using conventional phase
Extended Locking Range DTL using Chaos differentiation, thus avoiding noise enhancement, which is a
limiting factor at low SNR, and also does not utilize
Technique periodical preambles, which have undesired emissions and
Qassim Nasir, Saleh Al-Araji probability of false acquisition."
"The main objective of this paper is to broaden the lock
range of the Digital Tanlock Loop (DTL) by extending the
stable behaviour of the loop to larger control parameter by
incorporating a feedback loop. The tendency of the DTL to
diverge is measured and fed back as a form of linear
stabilization. Using this technique the loop lock range is
extended. The lock range extension has been confirmed
through the use of bifurcation diagram, and Lyapunov
exponent."

Tencon 2005 Page 88

 2C-09
Session Title: Orthogonal Frequency Divison Multiplex 2
Chairperson: Chandra Athaudage
Day: Wednesday
Date: 23rd November 2005
Time: 2:00PM – 3:40PM
Room: M8

2C-09.1 1568965152 2C-09.3 1568965550

A new interpolation method for channel GF(2^191) Elliptic Curve Processor Processor
estimation of OFDM systems using Montgomery Ladder and High Speed Finite
Shu-Hong Lee, H.L HUNG, Jyh-Horng Wen Field Arithmetic Unit
"Pilot-aided channel estimation of OFDM systems mostly Byeong-Yoon Choi
needs an interpolation technique to estimate the estimates "This paper describes hardware implementation of
of channel response on non-piloted subcarriers. Much GF(2191) elliptic curve cryptographic coprocessor which
research has been published concerning interpolation; supports 6 operations such as scalar multiplication(kP),
however, the performance of which is limited by either the Menezes-Vanstone(MV) elliptic curve cipher/decipher
number or the position of pilot subcarriers, such as cubic algorithms, point addition(P+Q), point doubling(2P), finite-
spline interpolation and DFT-based method. We propose a field multiplication/division. The ECC processor adopts the
new interpolation method derived from singular value Montgomery scalar multiplication scheme which main loop
decomposition (SVD). This method can effectively mitigate can be executed using only x-coordinate. Also it takes
the influences from the limitations mentioned above. In advantage of concurrent operations among arithmetic units,
consequence of simulation results, the proposed method such GF_ALU, GF_MUL, and GF_DIV to achieve high
has better performance than cubic spline interpolation cipher rates. It has about 68,000 gates and its simulated
under non-equidistant pilot pattern, in which DFT-based worst case delay time is about 7.8-ns under 0.35um CMOS
method could not work." technology. Because it has about 647 kbps cipher rate and
supports 8 operations, it can be efficiently applied to the
2C-09.2 1568965490 various elliptic curve cryptographic applications"
Low-Complexity Diversity Enhanced OFDM for
2C-09.4 1568965391
Wireless Communications
A Novel Wavelet Packet Transform Based
Kanagaratnam Sathananthan, Chandra Athaudage
Feature Extraction Algorithm For Image Texture
"We propose a low complexity frequency diversity Classification
technique for OFDM to improve its error performance in a
frequency-selective wireless channel. Recently, linear Selvan Srinivasan, Ramakrishnan Srinivasan
precoding has been proposed to achieve frequency A novel algorithm for extracting features of image textures
diversity (multipath diversity) in OFDM, which overcomes to facilitate better classification is described. The algorithm
the problem of channel nulls (deep fades in frequency- uses energy matrices at all subbands of the wavelet packet
domain). However, currently reported precoding techniques transform coefficients to extract the feature vector. The size
for OFDM have high computational complexity due to of feature vector is reduced by taking the weighted average
associated maximum-likelihood (ML) decoding or pseudo- of the feature vectors at all subbands. The classification is
inverse of large matrices (linear decoding). In the proposed performed by Euclidean classifier. The experimental results
diversity enhanced-OFDM (DE-OFDM), the decoding are presented to demonstrate the effectiveness of the
complexity is minimal as maximum ratio combining (MRC) proposed algorithm. The classification rate of the proposed
is incorporated at the receiver to generate a decision algorithm is found to be better than conventional wavelet
variable for each repeatedly transmitted symbol over packet signature (WPS) algorithm.
multiple subcarriers. Moreover, we show that DE-OFDM
does not cause any effective data rate (throughput) loss as
DE-OFDM using a higher-order modulation scheme (thus
the same data rate as conventional OFDM) can provide a
better error performance. Both analytical and simulation
results are provided to demonstrate the efficiency of DE-
OFDM in achieving frequency diversity benefit with low
receiver complexity in multipath (frequency-selective)
channels. DE-OFDM also outperforms the previously
proposed ICI selfcancellation schemes in a frequency-
selective fading channel with small frequency offset error."

Tencon 2005 Page 89

 2C-10
Session Title: Fuzzy Technique Applications 2
Chairperson: Kieran Orchard
Day: Wednesday
Date: 23rd November 2005
Time: 2:00PM – 3:40PM
Room: M10

2C-10.1 1568965434 table, hence a high-performance, cost-effective real-time

maximum power point tracking can be simply realized."
Fuzzy Logic Based Error Concealment Algorithm
for H.264 Video Transmission 2C-10.4 1568965200
Pei-Jun Lee, Ming-Long Lin Unmanned Light Craft Balance System Using
"In this paper, a new error concealment algorithm is Fuzzy Logic Control
proposed for the coding standard H.264. The algorithm
consists of a block size determination step to determine the
Liqiong Tang, Kieran Orchard, Alex Wishart
size type of the lost block and a motion vector recovery step "This paper presents an approach to the control of an
to find the lost motion vector. The former step uses fuzzy unmanned light aircraft. The light craft has similar properties
logic method to select the size type of the lost block from to a helicopter. It is fundamentally unstable and with highly
the current frame. In the latter step, the optimal motion nonlinear dynamics. A control system is required to control,
vector for the lost block is chosen from multiple previous compensate, and balance the craft when there is a change
reference frames with the minimum value of the side match on load capacities, disturbances, and environmental
distortion. The proposed algorithm determines the most conditions. A fuzzy logic control system based on a
correct mode for the lost block, and save much more microcontroller was developed to realize a light craft flight
computation time for motion vector recovery. Experimental controller."
results show that the proposed algorithm achieves 0.5~3 dB
improvement than conventional VM method." 2C-10.5 1568965210

2C-10.2 1568965565 Real-Time Control of a Binary Distillation Column

Using HGA Fuzzy Supervisory PI Controllers
2 to 16 GHz Microwave Line-of-Sight Path Loss
Prediction on Urban street by Fuzzy Logic KIATKAJOHN WORAPRADYA, Suvalai
Models Pratishthananda
"In this paper, an optimum fuzzy supervisory PI controller
SUPACHAI PHAIBOON using hierarchical genetic algorithms is developed and
"This paper proposes a method to model the path loss implemented for controlling the top and bottom product
characteristics on an urban street in microwave band. We quality of a nonlinear, multi-input multi-output binary
applied the concept of fuzzy logic to predict an unknown distillation column when the disturbance enters the column
path loss from a set of know path loss. The input fuzzy sets in the form of the changes in feed flow rate. Two
were classified into five sets namely, distance between conventional PI controllers, one for the bottom product
transmitter and receiver, frequency, time of day, composition and another for the top product composition,
transmitting antenna height and receiving antenna height. are used together in the control scheme. Hierarchical
These inputs are then inferenced into output path loss via genetic algorithms are used to derive the optimal number
linguistic rules which were trained by measurement. To and shape of membership functions and fuzzy rules of a
check the proposed model, we compared the fuzzy fuzzy supervisory system that adapts the parameters of the
prediction with an another measurement. The results shown PI controllers. The real-time implementation results show
that the fuzzy logic models provided a better prediction." the effectiveness of the proposed method."

2C-10.3 1568964662
Grid-connected Photovoltaic System with
Maximum Power Point Tracking using Self-
Organizing Fuzzy Logic Controller
Noppadol Khaehintung, Phaophak Sirisuk, Anatawat
Kunakorn
"This paper presents the design of a controller for the
maximum power point tracking of a grid-connected
photovoltaic energy conversion system. A boost converter
is used in the system to deliver the output from the solar
array to DC-AC inverter, and feed the power into the AC
grid. A self-organizing fuzzy logic controller is introduced for
the tracking algorithm. The duty ratio for the operation of the
boost converter is optimally adjusted in such a way that the
maximum power point, which normally varies according to
the environment, can be achieved. We demonstrate via the
simulation results that our proposed technique outperforms
over the conventional fuzzy logic controller in terms of
tracking speed and transient response. Furthermore, the
algorithm implementation can be done using a look-up

Tencon 2005 Page 90

 2C-11
Session Title: ATNAC Poster Session 2
Chairperson: Richard Harris
Day: Wednesday
Date: 23rd November 2005
Time: 2:00PM – 3:40PM
Room: M1

2C-11.1 1568965407 time of 27 min/year, which is smaller than that of the

copper-based access network."
An Architecture for Automated Network Control
of QoS over Consumer Broadband Links 2C-11.4 1568963144
Lawrence Stewart, Grenville Armitage, Philip Branch, Mobility Prediction with Direction Tracking on
Sebastian Zander Dynamic Source Routing
"This paper proposes a novel architecture for meeting Klangjai Sithitavorn, Bin Qiu
Quality of Service (QoS) requirements of real-time traffic
across consumer broadband links. In our approach to "Mobility prediction is one the most efficient techniques
network QoS provision, QoS signalling is moved from the used to calculate lifetime estimation of wireless links (link
application layer to the network layer. Network servers lifetime-LLT) for a more reliable path selection. To
automatically identify traffic that might benefit from QoS and incorporate a mobility prediction model into ad hoc routing
then trigger the provisioning of QoS by signalling network protocols, an updating scheme should be used to maintain
elements such as access routers. This approach removes prediction accuracy especially for those source initiated on-
the need for the application to signal to the network its demand routing protocols where LLT values are
explicit QoS requirements, making applications easier to continuously changing after the route discovery phase. In
develop and more portable. It also enables QoS provision this paper, a light-weight direction tracking scheme is
for legacy applications for which there is limited opportunity introduced to enhance performance of the mobility
to include explicit end-host signalling protocols. The paper prediction based routing by reassuring the accuracy of LLT
develops the architecture required to realize the approach used for path selection process. The simulation results
and discusses the underlying techniques." show an improvement in terms of average packet latency
with minimum impact on number of control messages
2C-11.2 1568965424 utilized."

A QoS Provisioning Technique in a Residential 2C-11.5 1568965293

Gateway using Heterogeneous Network Access
Technology Bluetooth Network Security: A New Approach to
Secure Scatternet Formation
Lee Hyun Jin, Jae-Hyun Kim, Youn-Kwae Jeong,
Chunglae Cho Sanif Sentosa Liong, Payam Mamaani Barnaghi
"We propose a packet scheduling algorithm for an efficient "The Bluetooth technology, designed and developed by the
transmission in a residential gateway used in the home Bluetooth Special Interest Group (SIG), which comprises of
network. The packet scheduling algorithm is modeled by the Piconet and Scatternet. Scatternet is a collection of several
hybrid method that is a QoS technique based on the piconets where each piconet consists of at least two
parameterized QoS method and the prioritized QoS devices in different roles (i.e. master and slave). There are
method. We applied the prioritized QoS technique to various protocols for Bluetooth scatternet formation. In this
Earliest Deadline First scheme. We proposed a QoS paper, we study some of the common formation protocols
parameter mapping scheme that can be used in the home such as BlueTrees, BlueNet, and BlueStars. Due to the
network using the heterogeneous access technology. We open-air media, higher security mechanisms are needed to
also proposed the fairness scheme for IEEE 1394. The provide secure communication among the nodes in the
access technologies such as IEEE 802.11 WLAN, IEEE scatternet. To provide a secure communication between
1394, and Ethernet are considered for the QoS algorithm two parties, encryption mechanisms are used, which require
mapping scheme. Finally, we evaluate the performance of secret key for every pair. The focus of the suggested
the proposed QoS provisioning technique by using the method is the scatternet communication security and in
simulation. The simulation results show that the proposed particular the secret key exchange. In this paper, we
schemes are a good candidate for the home network." propose a mechanism for the key agreement procedure
through a secure scatternet formation protocol."
2C-11.3 1568964790
2C-11.6 1568965317
Ethernet PON or WDM PON: A Comparison of
The Implementation of Security Message
Cost and Reliability
Protocol for PDA PUSH Service
AN TRAN, Chang-Joon Chae, Rod Tucker
Lee Jeong Kyoon, Ki Young Lee,
"We report a comparison of cost and reliability of two optical
broadband access technologies namely Ethernet passive "In this Paper suggests the method to implement PUSH
optical network (EPON) and WDM PON (WPON), taking Service in 2.5Generation mobile device which can't support
into account two protection schemes and two deployment voice and data service simultaneously. And If this PUSH
scenarios. We derive simple network models and a figure of Service is implemented and serviced well, the suggested
merit for network cost and reliability evaluation. We find that available security protocol can be implemented using
with a take rate ≥ 60%, WPON technology provides current common algorithm. The PUSH Service use
the lowest cost per subscriber per Mbit/s. Moreover, with full SMS(short message service) in order to connect to wired
protection, the PON architectures achieve a low mean down network in off-line client device to data communication.
After receiving SMS message, client device process the
SMS message and create data channel through

Tencon 2005 Page 91

RAS(remote access service), then data of server can be 2C-11.9 1568964641
pushed to clients. Also This Paper implements security
protocol considering feature of PUSH Service and mobile Performance Evaluation of an Enhanced GDCF
device below, First, using secrete nonce table to re-duce Duk Hun Kim, Seung-Hyuk Choi, Myoung-Hee Jung,
the number of transmission Second, adapting two-way Min Young Chung, Tae-Jin Lee, Yutae Lee
transmission method using two channels of SMS and data.
In order to implement the security protocol, This Paper uses "IEEE 802.11 is one of the most interesting technology in
SEED and MD5 algorithm with fast process time and little wireless local area networks. In this paper, we investigate
memory space, and also uses RSA which has some performance of IEEE 802.11 DCF (Distributed Coordination
features of certification and encryption for safe security Function) and an enhanced GDCF (gentle DCF) in terms of
protocol. Because of using two-way method, suggested throughput and MAC delay under normal traffic condition.
security protocol can support safe transmission on each We also present an analytical model for the enhanced
communication line, also using security nonce table, the GDCF MAC delay under saturation condition. From the
protocol can reduce number of transmission in order to results, the performance of the enhanced GDCF is nearly
exchange safe session key, so number of bits of encryption same as that of conventional DCF for light traffic load, but
increased" the enhanced GDCF yields better performance than the
conventional DCF for heavy traffic load."
2C-11.7 1568965403
2C-11.11 1568963010
Transparent IP layer Interception on Enterprise
802.11b/g Networks A New Cellular Search Algorithm For Motion
Estimation
Warren Harrop, Grenville Armitage
Jiann-Der Lee
"Many enterprise sites utilise 802.11b/g technology to
create an untrusted access network sitting outside their "A fast block-matching motion estimation algorithm for video
protected institutional IP network, with internal access compression is desired in real-time video conference
allowed only through an IP-layer virtual private network applications. Three-step search (TSS) and other fast search
(VPN) gateway. Often such networks do not implement link algorithms, such as diamond search (DS), cellular search
layer security because of the known weaknesses of the (CS), new cross diamond search (NCDS), fast cellular
IEEE's wired equivalent privacy (WEP). This results in a search (FCS) have been proposed to reduce the
wireless network on which arbitrary people can establish computation time instead of the full-search block-matching
themselves as hosts with arbitrary IP addresses. Although algorithm in video compression applications. To reduce
the enterprise IP network is protected by the VPN gateway, more computational complexity and maintain the quality of
users of the wireless network can become victims of the compressed video, a novel algorithm, based on the
unscrupulous (or accidental) interception of their IP spirit of CS and named new cellular search algorithm (NCS)
communication. Common Windows laptop (mis-) with different switching strategy criteria, is proposed. The
configurations often try and establish communications experimental results are shown that the proposed NCS is
through a default gateway on the 192.168/16 network. superior to some fast search algorithms."
Anyone could configure another host as this default
gateway on the enterprise 802.11b/g network, and thus
hijack a visitor's network connection without the visitor even
realising. In this paper we test and confirm the plausibility of
this attack in a University wireless LAN and present results
from real world data confirming the existence of users
failing to reconfigure their visiting host and attempting to
connect via possible malicious gateways."

2C-11.8 1568965193
Optimizing Network Control Traffic in Resource
Constrained MANETs
Muhammad Iqbal
"The exchange of Network State Beacons (NSBs) is crucial
to monitoring the dynamic state of MANETs like sensor
networks. The rate of beacon exchange (FX) and the
network state define both the time and nature of a proactive
action to reconfigure the network in order to combat
network performance degradation at a time of crisis. It is
thus essential to select the FX within optimized bounds, so
that minimal control traffic is incurred due to state
maintenance and reconfiguration activities. This paper
presents a novel distributed model that selects optimized
bounds for FX selection and adapts dynamically to the load
profile of the network for energy efficient monitoring and
proactive reconfiguration."

Tencon 2005 Page 92

 2C-12
Session Title: Tencon Poster Session 3
Chairperson: Catherine Edwards
Day: Wednesday
Date: 23rd November 2005
Time: 2:00PM – 3:40PM
Room: M2

2C-12.1 1568965396 2C-12.4 1568965277

An Experimental Study of Random Noise A Parallel Estimator with LMS Adaptation for Fast
Characteristics in a Power Converter Fading Channels
Anawach Sangswang Shingo Oikawa, Yusuke Tsuda, Tetsuya Shimamura
This paper focuses on the quantification of existing random "We propose a novel channel estimation technique for fast
noise in a switching dc-dc boost converter. Theoretical fading channels. The proposed scheme supposes multiple
background regarding to the probability density of random linear transversal filters, which are constructed in a parallel
noise is provided. An experimental study is performed to structure. The estimator coefficients for the proposed
measure and characterize random noise on system estimation scheme are adapted by the least mean square
parameters of a boost converter. The converter analyzed is (LMS) algorithm. To obtain a further performance
the commonly used current-mode controlled boost improvement of the proposed scheme, a technique to adjust
converter. Experimental results show that the the step-size for the LMS is also introduced. Computer
characteristics of random noise in the converter are simulation results show that the proposed method provides
gaussian distributed. The quantified results of uncertainties a significant improvement related to the conventional LMS
and the probability density are included in the dynamics of estimator in high fade rate conditions."
the converter as load and switching time uncertainties. This
observation introduces the possibility of sensitivity analysis 2C-12.5 1568964019
at a design stage in which more robust design can be
achieved. Dynamic Spectrum Allocation Using Channel
Power in CDMA Networks
2C-12.2 1568965194 Jinyeol Roh, Youngnam Han
PWM Inverter IP for High-Fidelity Audio This paper presents and evaluates the performance of a
Applications temporal dynamic spectrum allocation (DSA) scheme based
on the CDMA networks as an alternative to the fixed
Maeum Lee, Kichul Kim, Rinchul Kim spectrum allocation (FSA) methods. The paper considers
"This paper presents a pulse width modulation (PWM) the spectrum allocation between two code division multiple
inverter intellectual property (IP) for high-fidelity audio access (CDMA) systems with different time-varying traffic
applications. PWM inverters are more power efficient than pattern. The traffic pattern is based on the real CDMA traffic
analog amplifiers and can be designed in smaller size than data from Daegu urban area in Korea. A proposed
analog amplifiers. A high performance PWM inverter algorithm is investigated by the simulation based on CDMA
consists of several modules including over sampler, cross call processing and shows that can produce significant
point deriver, noise shaper and PWM generator. These gains for the radio access network (RAN) in DSA.
modules were usually implemented by software on digital
signal processing (DSP) processors. This paper shows how
to design these modules in hardware. An architecture was
implemented with VHDL. The design can be used as an IP
on FPGA and SoPC for high-fidelity audio applications."

2C-12.3 1568965159
Double-Pass Discrete Raman Amplifier with FBG
for Pump Power Optimization and Improved
Performance
Mas Izyani Md Ali, Muhammad Zamzuri Abdul Kadir,
Asmahanim Ahmad, Mohd Adzir Mahdi
"We present a report on the experimental validation of
double-pass discrete Raman amplifier based on dispersion
compensating fiber and high reflection fiber Bragg grating at
pump wavelength. The pump light at 1435 nm is launched
in counter-directional to the incoming input signal. The
double-pass direction of the signal is achieved through a
fiber loop mirror constructed by a circulator. We proved that
with the proposed design, it requires only 55% of the pump
power to achieve the same Raman gain compared to the
typical single-pass Raman amplifier."

Tencon 2005 Page 93

2C-12.6 1568965516 2C-12.8 1568965480
Performance Evaluation of CDMA Systems Based Hard Real Time Communication Using the
on User Location CANIOP Protocol
Seung Sik Choi Tarek GUESMI
"In order to support high data-rate service in an integrated "Distributed computer control in complex embedded
CDMA system, it is required to allocate resources for data systems oversees diverse electronic control units
users with high spectral efficiency. In this paper, we analyze connecting hundreds or thousands of analogue and digital
the performance of two resource allocation schemes such sensors and actuators. These systems are also called today
as a simultaneous resource allocation (SRA) scheme and a sensors networks. In an emrging class of open distributed
throughput maximization(TM) scheme. In the throughput real time and embedded (DRE) systems with stringent but
maximization scheme, the throughput of data users can be dynamic QoS requirements, there is a need to propagate
maximized when one among data users is allowed to QoS parameters and enforce task QoS requirements
transmit information at any given time. To apply this across multiple endsystems in a way that is simultaneously
scheme to an voice/data CDMA system, resources are first efficient and adaptable. Consequently, this paper presents
allocated to voice users and then residual resources are an inter-ORB protocol, witch make efficient use of the
allocated to data users based on the throughput advantages of Controller Area Network (CAN) and maps
maximization strategy. For more practical approach, the the CAN priorities to a band of CORBA priorities. Our
transmission power of mobile stations is assumed to be definition of this protocol is bound closely to the features of
limited. Due to this assumption, the performance of two RT-CORBA, the OMG Data Acquisition from Industrial
resource allocation schemes is varied according to the Systems(DAIS) RFP and the CAN bus, well known as a
location of mobile stations. We also show the performance real-time underlying network . We give the criteria of the
under perfect power control and imperfect power control. In definition of the priority field and show how prioritization of
order to verify the analysis results, we perform the messages over the physical CAN network can be achieved,
simulation under the same environment. Simulation and when adopting the use of RT-CORBA scheduling service
analysis results show that the TM scheme has higher which implements a dynamic scheduling policy."
throughput than the SRA scheme under the conditions that
the mobile is located in a limited boundary and the variation 2C-12.9 1568965459
of E_b/N_ois minimized."
An FPGA-based Architecture for Block Matching
2C-12.7 1568965099 Motion Estimation Algorithm
High Linearity 0.18um CMOS Mixer with Current Kasturi Rangan, Manohar Reddy, Sivakumar Reddy
Draining Technique Vustikayala
Harikrishnan Ramiah "An FPGA-based architecture to achieve a real time
processing of full-search block matching motion estimation
"The drive for cost reduction has led to the use of CMOS algorithm is described. Although many ASICs for motion
technology for highly integrated radios. This paper presents estimation have been developed, either the chip complexity
a new CMOS active mixer topology with current draining is too high or the optimal accuracy was not achieved. Based
features, to reduce flicker noise effect. Mixing is achieved on the regular parallel design a real-time FPGA chip is
by exploiting two transconductors with cross coupled designed and it has been implemented using 16-processing
outputs, which are alternatingly activated by the current elements to meet the real time requirement. The design can
commutating CMOS switches. The architecture can operate achieve clock frequency up to 120 MHz. Its performance
at low supply voltage by the use of switches exclusively has been compared with other architectures reported in the
connected to supply voltages. Realized in 0.18mm CMOS literature and the results are encouraging."
technology and operating from 1.8V power supply, the
design consumes 3mW of power. The mixer exhibits a third 2C-12.10 1568964886
order input intercept point (IIP3) of better than 17dBm. An
overall noise figure of 18dB is achieved at 100kHz. The Dynamically Authorized Role-Based Access
mixer also shows an input dynamic range performance of Control for Grid Applications
5.85dBm, with 150.3dB carrier feedthrough suppression."
Hanbing Yao
"Grid computing is concerned with the sharing and
coordinated use of diverse resources in distributed ""virtual
organizations.?The heterogeneous, dynamic and multi-
domain nature of these environments introduces
challenging security issues that demand new technical
approaches. Despite the recent advances in access control
approaches applicable to Grid computing, there remain
issues that impede the development of effective access
control models for Grid applications. Amongst them are the
lack of context-based models for access control, and
reliance on identity or capability-based access control
schemes. In this paper, we present an access control
scheme that resolve these issues, and propose a Dynamic
Role and Context-Based Access Control (RCBAC)
framework which extends the RBAC with context
constraints. The RCBAC mechanisms dynamically grant
and adapt permissions to users based on a set of
contextual information collected from the system and
user¯s environments, while retaining the advantages of
RBAC model. We also describe the implementation
architecture of RCBAC for the grid application."

Tencon 2005 Page 94

 2D-01
Session Title: Internet Traffic Management
Chairperson: Richard Harris
Day: Wednesday
Date: 23rd November 2005
Time: 4:10PM – 5:50PM
Room: M9

2D-01.1 1568964038 with defined bounds on packet delay and loss, packet
transfers need to be connection-oriented and the
Impact of IP Traffic on DPSK WDM Systems connections appropriately resourced. A virtual Internet
Xuetian Huang, Gang Chen connection with specified switching procedures, traffic
capacity, and quality outcomes would, in analogy to transfer
"Based on analytical method, we show that when IP traffic capabilities in ATM networks, be a defined Internet transfer
is loaded directly over differential phase shift keying (DPSK) capability. With appropriate transfer capabilities, an IP
wavelength-division-multiplexed (WDM) systems with network could work with, or take the place of, traditional
perfect dispersion compensation, the burstiness of IP STM and ATM networks, as well as well as exceed them in
packets has significant effects on the cross-phase capability. For instance, IP networking could be superior to
modulation (XPM) induced nonlinear phase noise, and STM and ATM in regard to multimedia communications."
therefore exerts an influence on the packet-error probability
performance of the systems. We show that a small IP traffic 2D-01.4 1568965135
can reduce the XPM-induced nonlinear phase noise and
therefore is beneficial for a low packet-error probability." OptiFlow - A Capacity Management Tool
Robert Suryasaputra, Alexander Kist, Richard Harris
2D-01.2 1568964684
"Intra-domain traffic engineering for routing protocols such
Comparing different approaches to the use of as OSPF (Open Shortest Path First) or Intermediate
DiffServ on the Internet System-Intermediate System (IS-IS) can be performed by
Don McNickle, Ronald Addie finding a judicious set of weights to allocate to the links of
the network. Unfortunately, methods proposed in the
"The existing Internet appears to be close to providing good literature for finding those weights may require a significant
quality service for a very wide range of services, probably computational effort. By way of contrast, weight setting
because the TCP protocols aim to achieve, and come close approaches based on Linear Programming can be shown in
to achieving, fair queueing, or processor sharing. The this paper to find suitable weights in the order of seconds.
DiffServ architecture aims to do better than this by providing Prior to this time, it was necessary to determine appropriate
different performance standards for different classes of weight settings in an off-line mode. From the results
service. The natural way to apply DiffServ is to allocate presented in this paper it can be demonstrated that
classes in accordance with the urgency or priority of the solutions can be located for this problem in a matter of
requests. However, another approach is to use DiffServ to seconds. This makes it possible to perform traffic
allocate service classes according to the size of the engineering for short term link overloads in real-time mode.
requests the concept of ""size"" can be defined in a variety The performance of this methodology has been verified by
of ways: total bytes in a flow, rate of a flow, or by a series of using simulation based on the well-known performance tool,
token buckets. We use simple queueing models to ns-2. The technique described in this paper is being
investigate how much improvement in performance could integrated into an optimisation module of a network capacity
be obtained by implementing this service discipline, and management tool called OptiFlow."
whether, as a consequence, it is unnecessary and perhaps
even dangerous to assign classes of service in accordance 2D-01.5 1568965320
with the type of service requested. The results suggest that
shortest job first offers considerable advantages over Traffic Aggregation in Hybrid Ring-Star Optical
simple processor sharing. Thus in spite of the difficulties of Networks
identifying the size of flows it may be worthwhile to consider
how something like shortest job first can be implemented.
Goutam Das
On the other hand it appears that priority queue strategies "The increasing demand of future Internet traffic requires
are risky, in that the marginal advantages gained by those optical traffic grooming technology to reduce the complexity
jobs that benefit from the priority scheme are very small, of the network. A network with optical cross connects and
and the majority of jobs can expect to suffer worse fibers has been considered as one potential architecture to
response times. In fact random priorities, used here as a achieve this goal. The design of this type of architecture
model for priorities based on characteristics not necessarily involves determining the number of network components
dependent on job length, such as urgency, produces almost and the interconnection topology. In this paper we show
the same deterioration in service as a scheme which might that a hybrid ring-star topology and waveband grooming
well be expected to be at the lower bound longest job first." can help to reduce the network complexity. This helps us to
minimize the number of nodes, and the number of ports
2D-01.3 1568965124 required in each node. We develop an analytical method for
the design of hybrid ring-star topology to reduce the capital
Internet Transfer Capabilities for Time Critical cost of the overall optical network. We evaluate the
Communications performance of our method in terms of the number of
Zigmantas Budrikis, Guven Mercankosk, John network elements required for a national backbone
network."
Siliquini, William Edwards
"Some time critical communications, e.g. telephony, can
frequently be conducted reasonably satisfactorily over the
Internet, even when packets are transferred in
connectionless mode. However, for any assured service

Tencon 2005 Page 95

 2D-02
Session Title: Multimedia 2
Chairperson: Sanjay Bose
Day: Wednesday
Date: 23rd November 2005
Time: 4:10PM – 5:50PM
Room: M7

2D-02.1 1568965318 2D-02.4 1568965416

Implementation and Usability of User Interfaces Design and Implementation of Stream
for Quality of Service Management Partitioning Strategies in a Distributed
Mladen Georgievski, Nalin Sharda Multimedia System
"This paper presents a comparative study of two methods Xiaorong Li, Bharadwaj Veeravalli
for implementing the Quality Cost Temporal Triangle "In this paper, we address the issue of rendering network-
(QCTT) user interface for dynamic QoS Management of based multimedia services under the guarantee of access
Networked Multimedia Systems. Two options investigated delay. We propose a novel strategy, referred to as Window-
are called: the Cartesian Coordinate System (CCS) and the assisted Video Partitioning (WAVP), for rendering cost-
Triangular Fractal System (TFS). We carried out a effective multimedia services. To minimize the service cost
comparative usability study for the graphical user interfaces and maximize the number of requests that can be
(GUI) implemented for the same. The GUI provides two successfully served under resources constraints (cache
methods of controlling the QoS, namely Pivot-Point and capacity and link bandwidth), our WAVP strategy partitions
Three Sliders control methods. Overall, the Pivot-Point videos into multiple portions and delivers them by adaptive
control method was rated higher by the participants for both schedule windows. We precisely implement our work for
CCS and TFS implementations." testing in real-life situation, and analyze the performance
under several influencing parameters such as the
2D-02.2 1568965387 availability of link bandwidth, cache capacity, and the bound
of clients' access delay. Our results show that the proposed
Investigation on the Content Popularity method can significantly reduce the service cost and
Distribution under K-Transformation in achieve a high acceptance ratio under the constraints of
Streaming Applications network resources."
Yang Zongkai, Tai Wang, Xu Du, Wei Liu, Jiang Yu
2D-02.5 1568965498
"An accurate understanding on the content popularity
distribution of media objects in streaming applications is DHT Based Collaborative Multimedia Streaming
required when designing efficient caching algorithms for the and Caching Service
streaming media. However, there is no empirical model to
describe this distribution. Our investigation shows that the
Zuoning Yin, Hai Jin
content popularity of a hot streaming media file exhibits a "Multimedia streaming and video on demand (VoD)
Zipf-like distribution under K-Transformation, after analyzing services are among those hottest applications over the
250 hot clips in 2 long-term traces. A specific function is Internet. Traditional client/server pattern is stumbled due to
proposed to describe this phenomenon. This result builds a the lack of scalability and limitation of service capacity.
foundation for designing intelligent caching algorithms." Peer-to-Peer paradigm has been introduced to solve these
issues. The key idea is that peers not only receive
2D-02.3 1568965402 multimedia stream but also serve multimedia data to others.
In this paper, we present a novel scheme that builds a
Improved Dynamic Multimedia Resource multimedia streaming and caching service on a distributed
Adaptation-based Peer-to-Peer Sys-tem through hash table (DHT) based overlay. We adopt DHT to
Locality-based Clustering and Service establish a decentralized caching service. The location
information of cached segments is distributed among
Letian Rong, Ian Burnett
different peers and the information about segments from the
"A dynamic P2P architecture based on MPEG-21 was same video is organized into an application-level multicast
proposed in our previous work to support resource group. We also propose a cache replace algorithm on each
adaptation/personalization according to the surround-ing peer. In our simulation, our design achieves good scalability
usage environment and user preferences. In this paper, we with a low service reject rate. Besides, good load balance is
improve the proposed system through two separate but also achieved as well as desired quality-of-service."
related modifications. Firstly, peers are clustered according
to registered geographic location information. Secondly,
based on that registered loca-tion information, a locality-
based service is introduced which allows users to search
services according to their geographic locations. The
service complements the proposed P2P architecture by
encouraging service providers to increase the uptime of
their devices and hence provide the spare computer power
for active adaptation of resources for low-end peers.
Simulation results show that the proposed approach
reduces download time and network delays while increasing
resource availability and download speed in the net-work."

Tencon 2005 Page 96

 2D-03
Session Title: Wireless LANS 2
Chairperson: Grenville Armitage
Day: Wednesday
Date: 23rd November 2005
Time: 4:10PM – 5:50PM
Room: P2

2D-03.1 1568965261 networks even if the topology varies, as is the case in a

mobile environment."
Evaluation of FAST TCP in Low-Speed DOCSIS-
based Access Network 2D-03.4 1568965519
Lachlan Andrew, Irena Atov, David Kennedy, Bartek Extending Cluster File Systems beyond Last
Wydrowski Miles
"There is strong evidence that the efficiency of the Internet Hann-Huei Chiou, Pan-Lung Tsai, Jiunn-Jye Lee,
is limited by its existing TCP congestion control system. A
replacement, FAST, has been shown to improve
Hsing-Fu Tung, Chun-Ying Huang, Chin-Laung Lei
performance in high-speed networks. In order to achieve "The growth of the bandwidth available in WAN links
widespread acceptance and standardisation, it must also be stimulates novel usage of traditional network systems. By
tested in environments more typical of the existing Internet. extending the boundary of cluster file systems to the
This paper experimentally evaluates the performance of customers' premise, it is now possible to provide home
FAST over a typical access link, with bandwidths of around users efficient, dependable, and responsive network
0.5-3\,Mbps. Links both using the DOCSIS cable modem storage. In this paper, we identify the primary issue of
medium access control (MAC) protocol and simple low rate network latency when implementing cluster file systems
links were investigated. It is shown that the random delay across the last miles, and propose a solution by replacing
introduced by MAC protocol of the cable modem does not the round-based data transmission protocol of Coda file
appear to interfere significantly with FAST's ability to set the system with a rate-based one. The performance evaluation
congestion window size to its target. However, the cable of the prototype system shows significant improvements of
modem does appear to introduce consistent additional both throughput and response time for file-transfer
delays when the link is highly, but not fully, utilised. These operations, especially under high network latency. The
unexplained delays mean that a larger congestion window result is a latency-resistant cluster file system."
is required, and must be taken into account when setting
FAST's parameters, notably the target queue size, alpha." 2D-03.5 1568965523

2D-03.2 1568965292 The Design and Implementation of a WSRF-

Compliant Information Service on Wireless Grids
Designing Rectangular Slot Loop Antenna for
WLAN Application Jianghu Yang, Gegang Peng, Changlai Huang, Ming
Li, Chuanshan Gao
Noppin Anantrasirichai
"On wireless grids users and resource providers are not
"This paper presents the design of rectangular microstrip only mobile, but also nomadic. Information services on
slot loop antenna that competently achieves dual frequency wireless grids have special requirements. Resource
operation by using a microstrip feed line. This antenna is information has to be updated when users and resource
designed at the resonance frequency 2.4 GHz and 5.2 GHz providers arrive and depart frequently. Meanwhile,
thereby covering the required frequency-bandwidth of 2.4- information services are also mobile, so the clients and
2.4836 GHz and 5.15-5.35 GHz, which are widely used in information sources should dynamically detect the
Wireless LAN (WLAN). The characteristics of antennas are existence and departure of information services. Information
proposed and analyzed for instance input impedance, S11 services on wireless girds may be not as powerful as those
parameter and far field patterns. The antenna is analyzed on wired grids, so information services are required to be
by Finite Difference Time Domain (FDTD) method which decentralized and light-weighted. In this paper we present a
can predict and analysis of the electromagnetic responses WSRF-compliant and P2P based information service
of complex problems." architecture to meet the special requirements on wireless
grids, and implement the architecture on Globus 4. We
2D-03.3 1568965381 extend the function of MDS4 to support wireless grid
resource discovery, query and monitoring. P2P mechanism
Price-based Max-Min Fair Rate Allocation in is employed to integrate information services to provide
Wireless Multi-hop Networks grid-wide resource information. The clients and information
Liansheng Tan, Xiaomei Zhang, Lachlan Andrew, sources dynamically probe nearest information services to
Moshe Zukerman query or register resource information. Some
implementation issues are discussed as well."
"When allocating rates in wireless multi-hop networks, one
difficulty comes from the so-called MAC (media access
control) constraint. To overcome this difficulty, this paper
proposes a price-based max-min fair rate allocation
scheme. Unlike existing schemes, our scheme is based on
the MaxNet principle using the maximum price of all nodes
along a transmission path to control the flows rate. Through
theoretical analyses and simulation results, we show that
the algorithm is able to meet the MAC constraint and to
achieve max-min fairness among multihop flows in wireless

Tencon 2005 Page 97

 2D-04
Session Title: Modulation and Coding 3
Chairperson: Katsuhiro Matsuoka
Day: Wednesday
Date: 23rd November 2005
Time: 4:10PM – 5:50PM
Room: P1

2D-04.1 1568965358 2D-04.4 1568965447

A new Modulation Scheme for GSM EDGE Equivalence of the Linear Symbol-level MMSE
Ari Viinikainen, Jarkko Vuori, Fei Liu and the Generalized RAKE
"The GSM system has evolved to the GSM EDGE system Girish Redekar, David Koilpillai
which increases the data rate with new modulation and Linear detection of the downlink signals of a direct-
coding schemes. This increase is achieved by improving sequence code division multiple-access (DS-CDMA)
the spectral efficiency of the system at the expense of system has seen significant research in recent years.
power efficiency. In this paper, we propose the use of a Notable amongst such efforts is the Generalized RAKE
spectral and power efficient modulation scheme, a (based on ML criterion) and the RAKE receiver based on
combination of frequency and phase modulation symbol-level MMSE. Both these approaches aim to
\emph{N}FSK/\emph{L}PSK, to further increase the data suppress interference even while exploiting diversity. We
rate of the current GSM EDGE system. We show that by show analytically that these two approaches are exactly
using the current EDGE channel coding and a 2FSK/8PSK equivalent.
modulation scheme the data rate can be increased 33 \%
with just 2 dB degradation in $E_b/N_0$ ratio with equal bit 2D-04.5 1568965525
error rate in the AWGN channel."
Bit-interleaving in fast FH/SSMA systems with
2D-04.2 1568965359 convolutional codes and turbo codes
Performance Analysis of Noncoherent UWB Yu-Sun Liu, Chun-Chih Yeh
Transceiver with PEWB under LOS Residential "Bit-interleaving is generally believed to be an effective
Channel Environment measure when convolutional codes and turbo codes are
employed with $M$-ary signaling. It is shown in this paper
Myoungoh Ki, Sungsoo Choi, Hui-Myoung Oh, Kwan that bit-interleaving should not be blithely employed in fast
Ho Kim FH/SSMA systems with $M$-ary FSK. Both theoretical
"Low rate ultra-wideband (LR-UWB) system requires a upper bounds and simulation results show that bit-
simple and cost-effective transceiver architecture to interleaving coded sequences before mapping increases
establish nodes in future sensor network. To satisfy this the bit error rates for convolutional codes, except when bit
requirements, the simple noncoherent on-off keying (OOK) error rates is very small. Simulation results also show that
UWB transceiver with the parallel energy window banks for turbo codes, both bit-interleaving data sequences in
(PEWB) giving high precision signal processing interface is encoders and bit-interleaving coded sequences before
proposed. The flexibility of the proposed system in multipath mapping drastically degrade the performance of the
channel environments is acquired with the pulse and bit system."
repetition method. To analyze the bit error rate (BER)
performance of this proposed system, a noise model in
receiver is derived with commonly used random variable
distribution, chi-square. BER of 10-5 under the line-of-sight
(LOS) channel environment is achieved with the integration
time of 80 ns and signal to noise ratio (SNR) of 16.3 dB."

42D-04.3 1568965373
A higher rate turbo equalization with multilevel
block-coded modulation
Katsuhiro Matsuoka, Eiji Okamoto, Yasunori Iwanami
"In mobile communications including the cellular phone, the
signal quality is often degraded due to multipath fading.
However, a demand for a high-data-rate and a large
capacity is growing even in the mobile communication. In
this paper, to achieve the higher data-rate mobile
communication, we propose a turbo equalization with
multilevel block-coded modulation (BCM). The turbo
equalization has an excellent ability of equalization and the
BCM can comparatively make the transmission rate high
and the calculation cost low. We combine these techniques
and investigate the transmission characteristics."

Tencon 2005 Page 98

 2D-05
Session Title: Time-Frequency Analsis and Wavelets 3
Chairperson: Peter O’Shea
Day: Wednesday
Date: 23rd November 2005
Time: 4:10PM – 5:50PM
Room: P3

2D-05.1 1568965054 2D-05.3 1568965188

An Approach to FPGA-based Time-Frequency The Application of Wavelet and Feature Vectors
Spectrogram by Real-Time Sweep Spectral to ECG Signals
Extraction Algorithm Aya Matsuyama, Mirjam Jonkman
Min-Chuan Lin, Guo-Ruey Tsai, Tsao-Kai Chang, "The Electrocardiogram (ECG) is one of the most
Shi-Sheng Chu commonly known biological signals, which are traditionally
"We have proposed an approach to build up an FPGA- analyzed in the time-domain by skilled physicians.
based time-frequency spectrogram for time-varying signal. However, pathological conditions may not always be
Instead of short time FFT algorithm, we have adopted real- obvious in the original time-domain signal. Fourier analysis
time sweep spectral extraction algorithm to reduce large transforms signals into frequency domain, but has the
memory usage. By fixing the sweep frequency range of disadvantage that time characteristics will become
local oscillator and the center frequency of Finite Impulse unobvious. Wavelet analysis, which provides both time and
Response (FIR) filter, we can obtain more effective usage frequency information, can overcome this limitation. In this
of FPGA hardware resource and stable spectrum analysis paper, Arrhythmia ECG signals were examined. There were
performance. The time-frequency scaling algorithm has two stages in analyzing ECG signals: feature extraction and
been used to guarantee the measurement of wide feature classification. To extract features from ECG signals,
frequency range (sub-Hz to MHz) for arbitrary input time- wavelet decomposition was first applied and feature vectors
varying signal. The high speed FPGA performance enables of normalized energy and entropy were constructed. Vector
us to build up a three dimension time-frequency quantisation technique was applied to these feature vectors
spectrogram for audio signal by accumulating the spectral to classify signals. The results showed that Normal Sinus
data against the evolution time." Rhythm ECGs and Arrhythmia ECGs composed different
clusters."
2D-05.2 1568965146
2D-05.4 1568965195
A Novel Technique for Merging of Multisensor
and Defocussed Images using Multiwavelets Prominent Region Based Object Recognition
using Complex Wavelets
Jawed Qumar, Ram Pachori
Pranam Janney, Sridhar Gangadharpalli, Sridhar
"We present a novel method for merging of images from Varadarajan
different sensors or defocussed images from same sensor.
In purposed method Discrete Multiwavelet transform Content based image retrieval system has evolved off late
(DMWT) has been used for decomposition. The advantages as one of the demanding research area in the current age
of DMWT is its sensitivity to different kinds of object of information technology. Precision while retrieving images
features in images and it does precise image analysis than has been a main criterion for evaluation of retrieval
wavelet multiresolution analysis. The purposed algorithm techniques. Higher precision could be attained when
assumed the images are registered from different sources. matching prominent regions becomes a part of the retrieval
This has become of interest due to increasing number of system. We have developed a new technique to detect
sensors in various applications such as medical, these prominent regions which would enhance the feature
surveillance, remote sensing. Fusion is a technique of characteristics of an image thereby leading to higher
extracting best of inputs and conveying to the output. precision in retrieval system
Fusion is achieved through the formation of fused pyramid
with the DMWT coefficients from the decomposed pyramids 2D-05.5 1568965341
of the source images. Coefficients are combined based on
importance of the coefficient in the bands. The fused image
A Novel Approach to Interest Point Detection
is obtained through DMWT reconstruction process. Results Based on Phase Congruency
obtained from purposed method significantly reduces Changzhu Wu, Qing Wang
distortion artifacts."
"In the paper, a novel method of interest points detection
based on phase congruency by using Log-Gabor wavelets
is proposed. Points of maximal phase congruency
correspond to all the different types of 2D features in an
image, including gray-level corners, line terminations, and a
variety of junctions within lines. Phase congruency is a
dimensionless measure invariant to the changes in image
brightness or contrast, hence it allows the use of universal
threshold values that can be applied over wide classes of
images. Moreover, different kinds of interest points can be
detected since it does not make any assumption on the
geometric structure of interest points in the proposed
method. Experimental results on synthetic and natural
scene images have shown that accuracy of detection and
localization of interest points can be achieved using the
proposed approach."

Tencon 2005 Page 99

 2D-06
Session Title: Signal Processing for Media Applications
Chairperson: fred harris
Day: Wednesday
Date: 23rd November 2005
Time: 4:10PM – 5:50PM
Room: M3

2D-06.1 1568964536 2D-06.3 1568965090

Determine Calibration Parameters with Satisfied Common and Separate Parameterizations of
Constraints for Coplanar Camera Calibration Vision Model Based Perceptual Post Filtering For
Kaset Sirisantisamrid, Takenobu Matsuura, Kitti Digital Colour Images
Tiraseth Chin Soon Tan, Hong Ren Wu
"Coplanar camera calibration is the process of determining "This paper presents a coder with post-filtering, based on a
a set of calibration parameters that describe the mapping vision model for digital colour images in YCrCb colour
between 2D control points and 2D image coordinates. spaces. The approach exploits intra-band and inter-
However, the optimal solution for coplanar calibration can orientation masking properties of the Human Visual System
be obtained by solving a set of constrained optimization (HVS) to identify, to estimate and to recover the amount of
problem. This paper presents a technique to determine the perceived visual information loss due to compression. By
calibration parameters for coplanar calibration, where a set incorporating the vision model into the decoding stage of
of orthogonality constraints is fully satisfied. The solving of the JPEG2000 verification model (VM8) with 2 variants of
camera calibration consists of two steps. The first step is an model parameterization of (a) model with common
estimating of calibration parameters using a closed form parameterization and, (b) model with separate
solution based on distortion free. In the second step, the parameterization, it yields comparable or superior visual
parameters estimated in first step are improved iteratively performance of image quality over that of JPEG2000 VM8
by nonlinear optimization method, where the constraints are with both the MSE and the CVIS options. Further subjective
enforced during optimization process. The computing of test results have shown that the 2 variants of model
parameters is directly performed on computer image array parameterizations mentioned earlier do not appear to
by assuming that scale factor is one. Consequently, the significantly outperform the other in terms of image quality."
image center can be simultaneously computed with other
parameters. The proposed technique is tested with a 2D-06.4 1568965278
synthetic image that known distortion. The result shows that
the proposed technique gives sufficient accuracy." A Visual Attention Model Based on DCT Domain
Yiwei Jiang
2D-06.2 1568965038
"In the last few years, computer vision research has had an
Scalable multiresolution Image Segmentation increased interest in modelling visual attention and a
and Its Application in Video Object Extraction number of computable models of attention have been
Algorithm developed. But most of these models are based on pixel
domain. Compressed domain technique deals with data
Fardin Akhlaghian Tab directly in compressed domain. Computational complexity is
"This paper presents a novel multiresolution image greatly reduced owing to avoiding the expensive inverse
segmentation method based on the discrete wavelet DCT computation required to convert values from frequency
transform and Markov Random Field (MRF) modelling. A domain to the pixel domain In this paper, we present a
major contribution of this work is to add spatial scalability to method of detecting Regions of Interests (ROIS) in DCT
the segmentation algorithm producing the same domain. We use the properties of DCT coefficients to get
segmentation pattern at different resolutions. This property low-level features of frames and combining feature maps
makes it applicable for scalable object-based wavelet into a unique saliency map. We use MPEG4 (Xvid) video
coding. The correlation between different resolutions of clips to test our algorithm. The experiment results show that
pyramid is considered by a multiresolution analysis which is our algorithm can detect ROIS from DCT domain
incorporated into the objective function of the MRF successfully."
segmentation algorithm. Allowing for smoothness terms in
the objective function at different resolutions improves
border smoothness and creates visually more pleasing
objects/regions, particularly at lower resolutions where
downsampling distortions are more visible. Application of
the spatial segmentation in video segmentation, compared
to traditional image/video object extraction algorithms,
produces more visually pleasing shape masks at different
resolutions which is applicable for object based video
wavelet coding. Moreover it allows for larger motion, better
noise tolerance and less computational complexity. In
addition to spatial scalability, the proposed algorithm
outperforms the standard image/video segmentation
algorithms, in both objective and subjective tests."

Tencon 2005 Page 100

2D-06.5 1568965303
Design and Performance Evaluation of Channel
Reservation Schemes for QoS Constrained
Multimedia Traffic
Warakorn Srichavengsup, Lunchakorn Wuttisittikulkij
"This paper considers the problem of contension resolution
problems for high bit-rate wireless communication systems
where the round trip propagation delay is relatively longer
than the slot transmission time. In this scenario, we found
that existing channel reservation schemes can not
guarantee the satisfaction to all users regarding different
Quality of Service (QoS). Thus in this paper, we propose 2
new channel reservation schemes which taken into
consideration priority service to more delay-sensitive
multimedia traffics, namely, Partial UNI and UNI+MT
schemes for multimedia traffics, an extension of our
recently proposed UNI scheme for single-class traffic. The
goal is to achieve the highest system performance and
enable each traffic types to meet their QoS requirements.
We evaluate the performance of each scheme by
mathematical analysis. The numerical results show that our
proposed schemes are effective in enabling different traffic
users to achieve the best successful rate possible in this
kind of environment."

Tencon 2005 Page 101

 2D-07
Session Title: Adaptive Systems and Signal Processing 3
Chairperson: Amin Sadik
Day: Wednesday
Date: 23rd November 2005
Time: 4:10PM – 5:50PM
Room: M6

2D-07.1 1568965299 2D-07.4 1568965225

Equalization of the Distorted Chrominance Signal Spectral Analysis of Delta and Sigma-Delta
using IIR Digital Filter with the Bernstein Modulators Using Linear FM Input Signals
Polynomial Lei Shang, Zahir Hussain
Vanvisa Chutchavong "In this paper we present a simple and intuitive way to
"This paper, we present the use of the Bernstein analyze different aspects of digital Delta and Sigma-Delta
Polynomials to obtain an IIR digital filter transfer function, modulators. The analysis is based on studying the spectrum
which can be applied to video equalizer for compensating of the signals at different stages of the modulator. Linear
the amplitude of chrominance signal in color television. In FM (LFM) signals are used as test signals for different
addition, experimental results are carried out by using the configurations. The evaluation of systems is simplified to
modulated 20T sine squared signal on the color sub carrier ""spectrum matching"" between the input signal sequence
4.43 MHz. It is shown here that the proposed equalizer can and the output sequence of the quantizer. ""Noise shaping""
either enhance or compress the chrominance signal in color is also analyzed for the verification of the linear model
television transmission." assumption."

2D-07.2 1568965455 2D-07.5 1568965509

State-Space Least Mean Square with Adaptive An Efficient Implementation of Generalized
Memory Sidelobe Canceler for Broadband Antenna Array
Signals
Mohammad Malik, Muhammad Salman
"In this paper, we present state-space least mean square
Mohammad Reza Sayyah Jahromi, Lal Godara
(SSLMS) algorithm with adaptive memory. SSLMS Generalized sidelobe canceler is an alternative structure to
incorporates linear time-varying state-space model of the a broadband antenna array processor using a tapped delay
underlying environment. Therefore, it exhibits a marked line filter with N taps. It consists of a fixed main beam to
improvement in tracking performance over the standard receive the signal from the look direction and a set of (L-1)
LMS and its known variants. Overall performance of auxiliary beams for an L element array. The weighted sum
SSLMS, however, depends on model uncertainty, presence of these beams is used to minimize the noise in the main
of external disturbances, time-varying nature of the beam requiring the inverse of a matrix of the order of (L-
observed signal and nonstationary behavior of the 1)N. This paper presents a method to estimate the weights
observation noise. The step size parameter plays an of the auxiliary beams in parallel requiring the inverse of a
important role in this context. However, because of lack of matrix of the order of (L-1) and thus reducing the
prior information of the uncertainties, it is difficult to suggest computational cost substantially. It is shown that the
an optimum value of the step size parameter beforehand. proposed method is more robust to look direction errors
As a logical approach to such problems, the step size compared to the original structure.
parameter is iteratively tuned by stochastic gradient method
so as to minimize the mean square value of the prediction
error."

2D-07.3 1568964926
A Single-Bit Digital DC-Blocker Using Ternary
Filtering
Amin Sadik, Zahir Hussain, Peter O'Shea
"In this paper, a single-bit dc-blocker is presented. It is
comprised of a ternary filtering stage preceded by a sigma-
delta modulator. Different techniques are used to generate
the ternary taps for hardware and performance optimization.
Both the input and the output of this dc-blocker are
assumed in single-bit format. The proposed dc-blocker can
easily be implemented with FPGA."

Tencon 2005 Page 102

 2D-08
Session Title: Medical Engineering and Agents
Chairperson: Eleazar Tortal
Day: Wednesday
Date: 23rd November 2005
Time: 4:10PM – 5:50PM
Room: M4

2D-08.1 1568965031 2D-08.4 1568960601

User Model to Design Adaptable Interfaces for A Novel Resource Advertisement and Discovery
Motor-Impaired Users Model for Ubiquitous Computing Environments
Pradipta Biswas Samit, Bhattacharya, Debasis using Mobile Agents
Samanta Ebrahim Bagheri, Mahmoud Naghibzadeh, Mohsen
"User modeling is an important strategy for designing Kahani, Faezeh Ensan
effective user interface. User model for able-bodied user is "Ubiquitous computing as a new branch in the field of
not always suitable for physically and mentally retarded distributed computing has received wide contribution from
people. There are some special characteristics of users and different researchers. In this novel computing model, vast
also of the interfaces, which have to be considered when range of computational and communication resources along
building a user model for disabled users. This paper with other types of services are gathered under a single
concerns about modeling motor-impaired users for system image based on certain predefined criteria.
developing personalized interfaces for those people. In the Ubiquitous computing and software mobile agents are both
proposed user model, main emphasis is given on making pursuing the aim of creating a collaborative problem solving
the user model application independent, clustering users environment. While the ultimate goals followed in each
according to their physical disability and cognitive level and technology differs from the conceptual view but they both
adapting the model with respect to individual user as well as share many ideas that can be mutually exchanged. As the
cluster of users." two fields advance they amusingly congregate. The
convergence allows faster progress since many shared
2D-08.2 1568965226 concepts have been to a great extent solved in the other
field. In the pervasive computing environment a balanced
Monitoring of the Coalesced Ice Ball in distribution of available resources among different tasks is
Cryosurgery an important task. Devising efficient resource discovery and
Eleazar Tortal, Ranjith Liyana-Pathirana, Vojislav Ilic, advertisement algorithms is therefore of much
Upul Gunawardana importantance. In this article we aim to provide two
algorithms for this problem using mobile agents. The
"By representation of collocation polynomials and proposed resource discovery algorithms use two different
visualisation of individual ice ball formation in multiple hierarchical and flat approaches. The simulations show a
probe-cryosurgical procedure, the radius of thermal good performance for both of the proposed models;
propagation from the cryoprobe tip centre can be however the hierarchical algorithm shows better results
represented. This serves as the basis for optimisation of based on some of the introduced factors."
cryoprobe placement to the targeted tissue or the surgeons
decision for the cessation of the freezing process during 2D-08.5 1568963071
cryosurgery."
Macroaction Synthesis for Agent Systems
2D-08.3 1568965553 Hiroaki Ueda, Takeshi Naraki, Kazunori Hosoda,
Design and Implement the Stimuli Capsule at In- Kenichi Takahashi, Tetsuhiro Miyahara
vitro Experiment "We present methods to synthesize macroactions for agent
Sang Woo, Ki-Won Yoon, Jyung-Hyun Lee, Y. K. environments and the methods are combined with SOS
algorithm that learns rules for agent's behavior using
Moon, Chulho Won, Tae Wan Kim
reinforcement learning and evolutionary computation. To
"Recently, the capsule endoscope has been developed to acquire useful macroactions, our methods use some kinds
observe image from the inside intestine. Such a capsule of numerical values evaluated in performing SOS algorithm,
endoscope does not have locomotion or hold by it self. To e.g. fitness values of actions or the number of transitions
get proper diagnosis and meditation, it is necessary to between rules. New macroactions generated by our
control capsule from outside to enable to guide locomotion. methods are fed back to SOS algorithm for learning rules.
The designed and implemented capsule has feasibility to By repeating macroaction synthesis and learning rules
improve moving speed of capsule or charge direction by alternately, rules for agent's behavior are acquired. The
electrical stimulus by electrical pad. When the electrical methods shown here have been implemented and some
stimulus contracts the small intestine, the capsule is move experimental results have been shown."
to backward. The implemented stimuli capsule always
monitors excessive energy transmit and automatically shut
down electrical stimuli to make more secure. Also the
stimuli capsule goes to stand by mode to save battery, and
reset by comparator circuit which is connected to two
stimulus pad. To verify the design and implement stimuli
capsule, in-vitro experiments are performed with pig¯s
small intestine."

Tencon 2005 Page 103

 2D-09
Session Title: Estimation and Modelling 2
Chairperson: Lei Shang
Day: Wednesday
Date: 23rd November 2005
Time: 4:10PM – 5:50PM
Room: M8

2D-09.1 1568965484 2D-09.4 1568965201

Layered Multicast: A Study of Loss Event Rate A Frequency-Selective Mobile Radio Channel
Estimation in a Low Level of Statistical with Hyperbolically Distributed Scatterers
Multiplexing Lei Shang, Zahir Hussain
Osman Ghazali, Suhaidi Hassan "In this paper we construct a frequency-selective
"Loss event rate is the main parameter of TCP-equation geometrical and time-variant wireless channel model with
model. In a low level of statistical multiplexing packet hyperbolically distributed scatterers for a macrocell mobile
dropped pattern at the bottleneck link is dependent on the environment. We verify this model against the statistical
behaviour of any flow that compete for the network properties such as the level-crossing rate (LCR), the
resource. Since TCP is the dominant protocol of the average duration of fades (ADF), the probability density
Internet, the behaviour of other protocols is greatly affected function (PDF), the cumulative distribution function (CDF),
by the volatility of TCP flows. In layered multicast and the autocorrelation functions (ACF) of this model.
communication, the protocols inability to synchronise packet Simulation results are verified against the analytical Wide
sequence number across multicast layers exaggerate this Sense Stationary Uncorrelated Scattering (WSSUS)
problem. This results in oscillatory loss event rate channel model."
estimation. Two-step loss event rate filtering has been
proposed to reduce the volatility of loss event rate filtering. 2D-09.5 1568964849
This paper presents a comparative study of loss event rate
estimation techniques for layered multicast protocol that has
General Calculation Formula on the Influence of
been conducted in a low level of statistical multiplexing." Extraction Steam Pressure Loss on Thermal
Economy of Unit
2D-09.2 1568964739 Zhang Xuelei
Analysis and Modeling of the Priority Serving "Extraction steam pressure loss will occur as extraction
Scheme (PSS) and the Efficient Utilisation of steam flows through the pipe and the heater. The thermal
Communication Channels economy of the unit is decreased by the occurrence of
extraction steam pressure loss. Based on the basic
Horace KING equations of thermodynamic system, the influence of drain
"The accuracy of queueing Network Models is affected by water enthalpy on thermal efficiency of unit is calculated in
the overall configuration of various communication protocls. the paper, and the law of formulas is found. It is proved by
In addition, various Network measurements and traffic the paper work that the influence of extraction steam
control techniques are seldom accounted for. In this paper pressure loss on thermal economy is actually the
some of the assumptions of PSS are considered to superposition of the influence of heater outlet water
formulate the source-to-destination over chains delay and enthalpy and drain water enthalpy on thermal economy.
the distribution of source-to-destination delay of a given Based on the paper work, the influence of extraction steam
Network chain" pressure loss on thermal economy of unit is built. The
method discussed in the paper is not only prone to be used,
2D-09.3 1568964896 but suitable for computer program. The method possesses
the universal property. The general calculation formula is
Equalization of Time Variant Multipath Channels proved to be right by an application example."
Using Channel Estimation Based Classification
Techniques
Yoshihiro Tsutsumi, Yusuke Tsuda, Tetsuya
Shimamura
"For the purpose of equalization of rapidly time variant
multipath channels, we derive novel adaptation schemes
based on the least mean square (LMS) algorithm. In two
methods proposed in this paper, instead of a coefficient
vector, a coefficient matrix is prepared for a linear
transversal equalizer(LTE). Based on the estimated channel
impulse response, the equalizer coefficients are selected
and updated by the LMS algorithm. Computer simulation
results show that with a small increase of computation, the
proposed methods outperform the amplitude banded LMS
equalizer recently addressed in high fade rate conditions."

Tencon 2005 Page 104

 2D-10
Session Title: IT Theory and Applications
Chairperson: Dixon Kwok
Day: Wednesday
Date: 23rd November 2005
Time: 4:10PM – 5:50PM
Room: M10

2D-10.1 1568964776 2D-10.4 1568965169

Discussion on Pervasive Computing Paradigm A database (DB) parallel processing for
Zhenjiang Miao computational physics
"Pervasive/ubiquitous computing is considered as the Dixon Kwok
fourth-generation of computing and the next generation of "To distribute a parallel process to several computer boxes
IT technology. As such an important computing technology, through a computational Grid, there are three fundamental
what its computing paradigm should be? For the present problems. They may be running on different platforms, for
network computing, its computing paradigms such as client- example, WinXP, Linux, etc. The process itself can be
server, message passing and the recent peer-to-peer are written in different programming languages, such as Fortran
well known. Are the network computing paradigms suitable 90 (an older version was Fortran 77), Ansi C, C++, and
and sufficient to the ""anytime, anywhere and any format, Java 2 (very common in Internet application). The last but
human-centered"" pervasive computing? This paper will not the least, if we distribute the lower level component of
analyzed the existing computing paradigms of the present the process, for example, in machine codes it is not
and past generation of computing technologies. Based on guaranteed that they will work on different CPU
the analysis and the unique feature of pervasive computing, architectures such as 32bit CPU and 64 bit CPU. We, here,
a new computing paradigm, Master-Server Paradigm is propose a simple parallel processing system that can be
proposed. In terms of the new paradigm, a Pervasive cross-platforms (different operating systems), cross-
Computing Based Multimodal Tele-home Healthcare languages/compilers (C++, Java, etc), and cross-machines
System is designed." (32bit and 64bit CPUs, etc) by using a database as an
interface of sharing and transferring information. A separate
2D-10.2 1568964246 databases table is created to synchronize the operations
between the two computers. The parallel processing system
Reliability and Sensitivity Analysis of Embedded was successfully tested by running a plasma simulation in
Systems with Modular Dynamic Fault Trees two WinXP computers."
Hsiang-Kai Lo, Chin-Yu Huang, Yung-Ruei Chang
2D-10.5 1568963656
"Fault trees theories have been used in years because they
can easily provide a concise representation of failure A Self-Distributing Software Environment with
behavior of general non-repairable fault-tolerant systems. Binary Interception for Multi-Projector Displays
But the defect of traditional fault trees is lack of accuracy
when modeling dynamic failure behavior of certain systems Naoki Hashimoto, Yoshihiko Ishida, Makoto Sato
with fault-recovery process. A solution to this problem is "In immersive projection displays which surround users with
called behavioral decomposition. A system will be divided high resolution images, a PC-Cluster is often used for
into several dynamic or static modules, and each module realistic and real-time image generation. However, it is quite
can be further analyzed using BDD or Markov Chains troublesome to develop applications which support parallel
separately. In this paper, we will show a decomposition processing on the PC-Cluster. It is also difficult to acquire
scheme that independent subtrees of a dynamic module are enough rendering performance because of the limited
detected and solved hierarchically for saving computation bandwidth of the PC-Cluster. Therefore, in this paper, we
time of solving Markov Chains without losing unacceptable propose a self-distributing software environment for multi-
accuracy when assessing components sensitivities. In the projector displays. This environment achieves minimum
end, we present our analyzing software toolkit that data communication based on a master-slave model. The
implements our enhanced methodology." communication mechanism is automatically applied to
target applications by intercepting APIs. Hence we can use
2D-10.3 1568964003 existent applications on the multi-projector displays without
any additional modifications."
An Energy Efficient Data Query Architecture for
Large Scale Sensor Networks
Ruay-Shiung Chang
"With the advance of micro-electronic and wireless
communication technology, deploying a large number of low
cost, small-sized sensor nodes over a vast area for
environment detection is becoming more practical. Setting
up gradients for delivering queries or data reports by
flooding the networks consumes a great quantity of energy
especially for a large scale sensor networks. In this paper,
we propose an energy efficient data query architecture. It
conserves energy by avoiding full network flooding in
setting up data forwarding path. Simulations show that the
proposed architecture is quite energy efficient."

Tencon 2005 Page 105

 2D-11
Session Title: Tencon Poster Session 1
Chairperson: Enn Vinnal
Day: Wednesday
Date: 23rd November 2005
Time: 4:10PM – 5:50PM
Room: M1

2D-11.1 1568964819 2D-11.4 1568965069

Design Patterns for Image Processing Algorithm Two-phases Parallel Neural Network Algorithm
Development on FPGAs Based on RPROP
Kim Gribbon, Donald Bailey, Christopher Johnston Xiong Zhongyang, Yufang Zhang, Ling Ou, Li Li
"FPGAs are often used as implementation platforms for "BP algorithm is widely used in the field of business
real-time image processing applications because their intelligence. Aimed at improving its relatively slow
structure allows them to exploit spatial and temporal convergence speed and its tendency to be trapped in local
parallelism. Such parallelization is subject to the processing minima easily, an improved two-phases parallel algorithm is
mode and hardware constraints including limited processing presented in this paper. The first parallel operation is to cast
time, limited access to data and limited resources of the about for minima area so as to avoid getting into local
system. These constraints often force the designer to optimal solution to some extent and accelerate
reformulate the software algorithm in the process of convergence, thereby reducing the number of epochs. The
mapping it to hardware. To aid in the process this paper second parallel operation is to shorten learning time. The
proposes the application of design patterns which embody experiments demonstrate that the improved two-phases
experience and through reuse provide tools for solving parallel BP algorithm has the better performance of
particular mapping problems. Issues involved in applying speedup."
design patterns in this manner are outlined and discussed."
2D-11.5 1568964630
2D-11.2 1568964884
Prime Implicant Computation Based on Adjacent
A study of High Capacity Image Steganographic Minterm Operations
System
Bo Wang, ZhenGuo Shi
Yuk Chung "A new method for generating prime implicants is presented
"With the development of mobile communication and in the paper. A minimization algorithm for incompletely
internet technologies, digital media can be transmitted specified functions based on the method is also given. The
conveniently over the network. In this process the procedure starts from a minterm, base minterm. By
algorithms used to protect secret messages during searching for its adjacent minterms from lower order to
transmission become an important issue. This paper higher order, a prime implicant is generated. Because
presents a new high capacity image steganograpic model adjacent minterm operations are simple for computers to
based on Discrete Cosine Transform (DCT), Vector process, the method is time saving. We have compared the
Quantisation (VQ) and Pseudo Random Number Generator proposed method to the previous works[1][10] over a set of
(PRNG). The proposed new system can embed more randomly produced inputs. The experiments show that the
information than traditional algorithms without compression. proposed method ia efficient for small size questins."
The recovered hiding image has high Peak Signal to Noise
Ratio (PSNR) value and good visual quality. This system is 2D-11.6 1568964876
robust in terms of both JPEG compression and other signal
processing attacks." Algebraic Semantics for Complete Interaction
Sequence
2D-11.3 1568965366 Lakshmi Narasimhan, Renyi Zhou, Shivakumar
3D Modeling from Using Spiral Cone-Beam Sastry
Trajectory "This paper describes the algebraic properties, which
Walita Narkbuakaew augment the usefulness of Complete Interaction Sequence
(CIS). Two new concepts, namely, event semantics and
"Tomographic imaging is a technique for exploration of a event-based architecture, are introduced. Examples are
cross-section of an inspected object without destruction. provided from several GUI testing scenarios in order to
Normally, the input data, known as the projections, are express the power of the algebraic formulation. Keywords:
gathered by repeatedly radiating coherent waveform Complete Interaction Sequence (CIS), Event algebra and
through the object in a number of viewpoints, and receiving Event-based architecture and GUI."
by an array of corresponding detector in the opposite
position. In this research, as a replacement of radiographs,
the series of photographs taken around the opaque object
under the ambient light is completely served as the
projections. The process is called photographic tomograph.
In this paper, we purpose the 3D shape extraction based on
using photographic tomography with spiral cone-beam
trajectory. We have demonstrated that the purposed
technique can rectify the problem of perspective occlusion
that occurs when using cone-beam trajectory"

Tencon 2005 Page 106

2D-11.7 1568965107 2D-11.11 1568965233
Spell Checker for Thai Document A New Practical MIMO Channel Estimation
Sarin Watcharabutsarakham Technique for IEEE802.16a: Performance
Analysis
"The objective of post-processing of OCR is to correct error
from OCR result. It is important to use a spell checker tool Micheal Drieberg, Kuan Min Yew, Varun Jeoti
to detect and to correct misspelled words. This paper "Both orthogonal frequency division multiplexing (OFDM)
proposes statistical method to find unexpectedly frequent and multi-input multi-output (MIMO) are key enabling
character sequences without relying on the dictionary. It is a technologies for future wireless communications systems.
flexible method to detect the out of vocabulary words. The For both these technologies to work optimally, the channel
corpus that used to create 3-grams is belongs to NECTEC between each pair of transmitter and receiver needs to be
(National Electronic and Computer Technology Center). estimated accurately. In an earlier work, we had proposed a
The result is 3-grams are selected to use as the spelling simple channel estimation method for a MIMO-OFDM
checker for Thai documents. The ArnThai software is OCR system based on IEEE802.16a (wireless Metropolitan Area
software, which used to evaluate the proposed technique." Network (MAN)). The channel estimates were obtained by
cyclically rotating the preamble in such a way that the time
2D-11.8 1568965310 domain channel impulse response (CIR) from each
Creation of a Grid environment and Grid Portals transmitter can be separated in time within one symbol
period. Results show that the overall system BER are
Bhawna Bansal, Seema Bawa nearly as good as the upper bound performance that is
"Grids are becoming platforms for high-performance and obtained with perfect channel state information (CSI) with
distributed computing. Grids benefit users by permitting only 0.5dB worst case degradation in performance."
them to access heterogeneous resources that are
distributed geographically and organizationally. It allows 2D-11.12 1568965527
users to execute compute intensive problems whose The Soft Encrypting Channel Based on Turbo-
computational requirements cannot be satisfied by a single code En-decoders for Wireless Data
machine. This paper describes how to create Grid
Transmission
environment and presents design and development details
of Grid portals. The development, as discussed in this Yang Xiao, Ying Zhao, Yu-ming Xie, Moon Ho Lee
paper is based on Globus Grid GT3. Keywords: Grid
"Turbo-code has been applied in many fields except for the
Portals, Portlets, Portal Requirement, Globus Grid GT3."
field of secret communication since its appearance. This
paper presents a method on designing soft encrypting
2D-11.9 1568965372 wireless channel based on turbo-code. The method utilizes
How Good is a UML Diagram? A Tool to Check It. existing wireless transmitting channel to realize secret
communication based on turbo-code en-decoders. This
Lavanya Konda, Bala Kishore, Mohanty paper also presents the encrypting method, which has the
Hrushikesha, R. K. Shyamasundar compositive elements of turbo-code en-decoders as secret
"UML is increasingly being accepted by practitioners for key to realize encryption. The structures of encrypting
developing object oriented programs. Different views of an system and decrypting system and the idea of encryption
application are specified by different types of UML are also discussed in the paper. This paper simulates on
diagrams. A UML diagram can be told good when it is the method with audio and image signal to check the effect
structurally consistent. It is shown that consistency checking of encryption and the validity of the method according to the
of UML diagrams can be made using OCL constraints. OCL designed program. This paper shows that the encrypting
constraints are translated to SQL statements and defined method can be extended into the design of channel codecs
on databases that store UML diagrams. In case of a for wireless data transmission."
structural inconsistency resulting a bad UML diagram , SQL
statement applicable to the case is applied on the table that
stores the diagram, generating a warning to designers. This
paper reports on a tool developed to check structural
consistency of UML diagrams. In the process we have
defined a generic method to translate OCL to SQL and also
a strategy for structural checking consistency of any well
defined diagrams."

2D-11.10 1568965042
Data Mining on Patient Data
Wensheng Gwsice, Wensheng Guo
"In this paper, we use machine learning schemes lR, FOIL,
InductH and C5.0 to generate decision trees and rules from
the examples in the medical dataset. The aim of our study
is to infer the patterns that can help doctors to identify,
recognize and predict the effect of the risk factors on the
long term subjective cure rates of patients who undergo
colposuspension. High test classification was sometimes
achieved. Our best results came when one learning method
suggested the preprocessing steps to be used for another
method."

Tencon 2005 Page 107

 2D-12
Session Title: Tencon Poster Session 4
Chairperson: Catherine Edwards
Day: Wednesday
Date: 23rd November 2005
Time: 4:10PM – 5:50PM
Room: M2

2D-12.1 1568965508 2D-12.4 1568965100

A Distributed Architecture for Power Network Ozone Sensors based on Layered SAW Devices
Fault Analysis with: InOx / SiNx / 36º YX LiTaO3 Structure
Ramesh Rayudu Alexandru Fechete, Wojtek Wlodarski, Kourosh
"This paper discusses a distributed diagnostic algorithm for Kalantar-Zadeh, Anthony Holland, Anurat Wisitsora-
fault analysis in power networks. DAPoN is a model-based at
algorithm that incorporates a hierarchical power network "Monitoring of air-quality it is a major task nowadays as the
representation and model. The structural model is a four- emission of pollutant gases increases particularly in urban
level representation with each level depicting a more areas. Ozone (O3) is a significant toxic pollutant and
complex organisation of components than the previous one presents health risks. We have developed ozone sensors
in the hierarchy. Each level is modelled by the object- using a Surface Acoustic Wave (SAW) based devices. The
oriented representation method. The functional sensors structure consists of a lithium tantalate
representation is also done in four levels where each level piezoelectric (LiTaO3) substrate with silicon nitride (SiNx)
contains the behavioural knowledge related to components intermediate layer deposited by R.F. magnetron sputtering
of that level in the structural model. The diagnostic and E-beam evaporation techniques. A 100 nm thin film of
algorithm of HiMoBFAD is designed to perform the fault indium oxide (InOx) deposited by R.F. magnetron sputtering
analysis in pre-diagnostic and diagnostic levels mainly to provide the selectivity towards O3. This paper presents a
ease the complex power network fault diagnosis. The comparative study of the sensors performance in terms of
diagnostic algorithm also contains separate algorithms to response time, recovery time and response magnitude as a
handle common-section, mal-functioning, mis-timing and function of operational temperature. Exceptionally large
multiple faults." frequency shift as high as 56 kHz was observed for O3
concentrations as low as 40 parts per billion (ppb) in air.
2D-12.2 1568963641 Microstructural characterization of the InOx thin films by
A Kind of Adaptive Genetic Algorithm and Atomic Force Microscopy (AFM) is also presented."
Application in Nonlinear Model Identification
2D-12.5 1568968746
Changliang Liu
Performance of a Text-Independent Remote
An adaptive genetic algorithm based on float-encoding is Speaker Recognition Algorithm over
presented and applied in nonlinear model identification. Communication Channels with Blind Equalisation
This algorithm is able to modify its own crossover rate and
mutation rate during the search according to the fitness Katrina Neville, Jusak Jusak, Zahir Hussain,
adaptively. The improvement can guarantee the colony Margaret Lech
multiplicity and the convergence. The simulation results of
“In this work we will present a study of the reliability of a
identifying a theoretical model and application to a real
well-known speaker recognition algorithm when using
object have proved that the adaptive algorithm leads to
speech sent over communication channels with Channel
significantly superior solutions with less computation time.
distortion and noise. The speech features used to test and
train this system are the Mel-Frequency Cepstral
2D-12.3 1568965487 Coefficients. For speaker recognition applications, channel
Boiler combustion control using the intelligent deformations can lead to serious errors in recognition if the
controllers speech is transmitted, making the algorithm unreliable for
usage in telephone banking or other applications requiring a
Yunhe Du, Jianchun Xing high level of security. We will study the performance and
Now many of the small scale power plants are controlled by reliability of this algorithm for text-independent speaker
the conventional analog instruments with poor performance recognition with speech sent over a communication
and reliability. It is an economic way to use the intelligent channel. We will be using blind equalisation techniques with
digital controller to upgrade the automatic control and QPSK modulation."
instrument system of these power plants. In this article the
intelligent feature of a new kind of digital controller is 2D-12.6 1568963021
described briefly and then its application in the boiler Analysis of Conversational Game Pidgin
combustion control is discussed. The controlled process is
Language
analyzed first and then a control strategy is designed to
obtain good control performance with the PID parameters Tarashankar Rudra
auto-tuned. The stable operation of this control system
"In this paper we analyze a new language called Game
more than one year proves that the design of this system is
Pidgin (Hall, 1966) Language (GPL) (Rudra et al, 2003;
successful.
Rudra et al, 2005) for speech interactive computer games.
We use two formulas namely entropy and perfect
information of Information Theory (Shannon, C., E., 1948)
to perform the analysis. We establish that GPL is an
efficient language for communication in real time speech
interactive systems."

Tencon 2005 Page 108

2D-12.7 1568965168 Javed Ahmed, M. N. Jafri, J. Ahmad
Compression and Enhancement of Digital "In this paper, we present a comprehensive study to design
Mammograms with Wavelet Footprints an artificial neural network (ANN) for tracking a target in an
image sequence. The proposed ANN architecture is a
Tatt Wei Ho, Varun Jeoti single-hidden-layer back-propagation neural network
"We propose a near lossless representation of digital (BPNN), in which the sigmoid and the linear activation
mammograms using the wavelet footprint basis, that can functions are used for its hidden and output layers,
simultaneously enhance the diagnostic features of the respectively. The features used for the input layer of the
mammogram and yield some compression. Our simulations BPNN are 4th level Daubechiess wavelet decomposition
on sample mammograms show that this representation can coefficients corresponding to the input image.
achieve upto a 4X reduction in the number of samples Performances of db1, db2, db3, and db4 wavelet features
required. The mammograms reconstructed from this are compared. The object, which is tracked for the purpose
representation have Peak Signal to Noise Ratios (PSNR) of of demonstration, is a specific airplane. However, the
around 39 to 40dB on average. A representation that is proposed ANN model can be trained to track any other
approximately 6X smaller can also be obtained without object of interest. The trained ANN has been simulated and
further degradation of the image quality if a simple region- tested on the training and testing datasets. The tracking
of-interest coding based on low order statistics of the error is analyzed with post-regression analysis tool, which
mammogram images are used with the wavelet footprint finds the correlation among the calculated coordinates and
representation. More significantly, important diagnostic the correct coordinates of the object in the image. The
features in the mammogram can be visually enhanced by a promising results of the presented computer simulation and
simple scaling of the footprint components of the analysis show that the proposed target tracking technique
mammogram." exploiting the power of ANN and wavelet transform is quite
plausible and significantly robust."
2D-12.8 1568965357
2D-12.11 1568965692
Multiresolution Image Alignment based on
Discrete Wavelet Transform Importance of Accuracy For Steady State
Performance Analysis of 3f Induction Motor
Petch Nantivatana
P. Giridhar Kini, P. N. Sreedhar, K.R. Varmah
"We introduce a multi-resolution image registration based
on using discrete wavelet transform. We first extract contour "This paper brings out the need for achieving accuracy in
from both images that we want to align. The extracted estimating the various operating parameters like motor
contours are then fitted with B-spline curve representation efficiency, losses, operating power factor etc in order to
to synthesize the new contours with equal number of point. analyze the steady state conditions in non-ideal situations,
The area parameter is used in the B-spline fitting to make with an analytical study. Core losses have been usually
the new generated curve immune to affine transformation. neglected in the analysis with the assumption that they are
Before representing the B-spline contour with discrete quite negligible in nature. Voltage variations have become a
wavelet transform, the problem of starting point of the common phenomenon as of late. As core losses are
contour needs to be handle. This can be done by computing dependent upon the supply voltage, it is therefore
the maximum curvature. The maximum curvature is necessary to include the core loss in the analysis, however
selected as the starting point. Once the starting points on small they may be. This example takes into account both
the contour have been established, the discrete wavelet overvoltage and undervoltage conditions, with upto 5%
transform is then recursively represented the contours until unbalance variations."
only a few points is remained. Due to the affine-invariant
properties of discrete wavelet transform, these points can 2D-12.12 1568964660
be used as landmark points for registering the transformed Optical Signal Processor for Multibeam array
contour with the original contour. The experiments have
shown that the purposed methods are robust and promising
antennas Using Inverse Piezoelectric Effect
even in the presence of noise." Elizabeth Caroline
"In this paper, we propose a novel optical signal processing
2D-12.9 1568965412 technique for a beam-forming network in cellular mobile
Visual Assessment for The Quantization Error in basestation. The inverse piezoelectric effect in PVF2 coated
Wavelet Based Monochrome Videos single mode fiber is employed to control the phase and
amplitude weights of each element of the smart antenna
Liming Mei, Hong Ren Wu array. The antenna weights are calculated for the given look
The investigation of the Discrete Wavelet Transform (DWT) direction and null directions using maximum directivity
based video coder is still undergoing in the literature. One algorithm. Beam formation is simulated for multiple users
of the open problems to be solved is the perception to the and beam steering is found to be satisfactory in a
quantization noise in different subbands in the DWT 120˚ sector."
domain. This is a critical issue for the development of a
better motion compensation (MC) scheme. An experiment 2D-12.13 1568965534
and relevant results analysis are presented in this paper to Adaptive LMS Ternary Filtering
address the above issue. Monochrome video sequences of
natural scenes are used in the experiment therefore the so- Amin Z Sadik, Zahir M Hussain
called masking effects can be taken into account in the
The unresolved problem of finding an adaptive LMS
decision of the sensitivity to the noise hidden in the DWT structure for ternary filters has substantially limited their
domain. The preliminary results show that the most
useability despite their major advantage of hardware
sensitive subbands are those in the lowest three resolution
simplicity. In this paper we tackle this challenging problem
levels under a five-levels decomposition scheme. The by introducing an algorithm that seems to be quite
further analysis proves that the distribution of the sensitivity
promising. Results were surprising as it appeared that the
to each individual subband has been shifted by the context
proposed ternary algorithm is comparable in performance to
of the video. the multi-bit Wiener LMS algorithm. We expect that this
approach will open the door for ternary systems to be ready
2D-12.10 1568965471 for replacing multi-bit signal processing systems.
Target Tracking in an Image Sequence Using
Wavelet Features and a Neural Network

Tencon 2005 Page 109

 3B-01
Session Title: Mobile and Wireless Networks
Chairperson: Moshe Zukerman
Day: Thursday
Date: 24th November 2005
Time: 9:30AM – 10:50AM
Room: M7

3B-01.1 1568974020 3B-01.4 1568966311

A New Quasi-Lumped Resonant Cell Concept in Perceptual Influences on Use of Fixed Lines and
Modern RF and Microwave Components Mobile Phones
Tsz Yin Yum Gabriele Hermansson, Anthony Saliba
"The quasi-lumped resonant cell concept provides a new "Not withstanding the increase in uptake of mobile phones
paradigm for modern RF front-end component designs in Australia, fixed line phones remain in extensive use. The
where simple, compactness, low cost and high performance Australian population is one of the highest users of fixed
are critical. This paper reviews some of the recent works in line phones in the world. In the absence of any models
this area which has been pursued by the authors research appropriate to this domain, a model for uptake and usage of
group at HKCityU. With a special pattern etched on the established technology, and in particular fixed lines is
transmission line, the devised resonant structures feature proposed. The findings of this study revealed that the
very simple, yet compact 1-D designs that offer very broad variables that influence drivers to use fixed lines differed
stop-band, and moderate slow-wave characteristics. from those that predict the use of mobile phones. The
Numerous passive and active RF/microwave components model also explores the contribution of a new proposed
are demonstrated to show the uniqueness of this idea." predictor variable towards the continued use of technology
Subjective Newness."
3B-01.2 1568964670
3B-01.5 1568965371
Modification of An Open Area Okumura-Hata
Propagation Model Suitable For Oman Wireless Grid Middleware Architecture for
Providing Reliable Services
Nazar Mohamed
"This paper examines the applicability of Okumura-Hata
Yong Hyuk Moon, Tran Minh Trung, Chan-Hyun
model in Oman in GSM frequency band. The study was Youn, Song Hyeon-Sun, Ji Jung-Yeop
carried out for open area only, since measurements "Currently, there are many requirements for the inter-
provided from OmanMobile were about the open areas. The connection and interoperability between mobile users and
mean square error (MSE) was calculated between static Grid area with a P2P-like manner. Due to the nature
measured path loss values and those predicated on basis of the mobile device and quality poor connection in wireless
of Okumura-Hata model for an open area [1 ]. The MSE is network, the behavior of mobile devices might be very
up to 6dB, which is an acceptable value for the signal unpredictable. Thus, we need well-organized system
prediction [1]. Therefore, the model gave a significant architecture to provide reliable Grid services for the mission
difference in an open area that allowed necessary changes critical application such as e-Health services. In this paper,
to be introduced in the model. That error was minimized by we introduce a new Mobile-to-Grid Gateway Replication
subtracting the calculated MSE (15.31dB) from the original (MGR) architecture that can satisfy those requirements. In
equation of open area for Okumura-Hata model. Modified MGR architecture we integrate Mobile-to-Grid Gateway
equation was also verified for another cell in an open area (MG) with Policy Based Quorum Resource Management
in Oman and gave acceptable results." (PQRM) [1][8] into a unity system. And we also propose a
Cost-constraint replication algorithm for calculating the
3B-01.3 1568965386 optimal number of replication gateways depending on
users?budget and requirements. Through the theoretical
Sensor Fusion Based Localization of a Mobile analysis and simulation, the results from the proposed
User in a Wireless Network scheme are measured and discussed."
Adrian Bishop, Pubudu Pathirana
"This paper applies sensor fusion to the localization
problem of a mobile user. We propose that the use of
direction of arrival (DOA) estimations along with received
signal strength measurements can increase the accuracy
and robustness of location estimations. The DOA
estimations are incapable of providing multi-dimensional
positioning alone, while signal strength methods are prone
to high uncertainties. A Robust Extended Kalman Filter
(REKF) is used to derive the state estimate of the mobile
user's position, and successfully track the mobile users with
less system complexity, as it requires measurements from
only one base station. Therefore, localization of mobile
users can be performed at the single base station.
Furthermore, the technique is robust against system
uncertainties caused by the inherent deterministic nature of
the mobility model. Through simulation, we show the
accuracy of our prediction algorithm and the simplicity of its
implementation."

Tencon 2005 Page 110

 3B-02
Session Title: Computer Networks
Chairperson: Radia Perlman
Day: Thursday
Date: 24th November 2005
Time: 9:30AM – 10:50AM
Room: P2

3B-02.1 1568963324 3B-02.3 1568964733

Merging Retrieval Results in Hybrid Peer-to-Peer Educating Nomadic Children: An Experiment with
Networks the Convergence of Technologies
Qian Zhang, Xuezhi Wen, Yu Sun Muriyankulangara Ananthakrishnan
"The state of the art of hybrid P2P-based information "EDUCATING NOMADIC CHILDREN: AN EXPERIMENT
retrieval is still at its infant stage and confronted with many WITH THE CONVERGENCE OF TECHNOLOGIES The
challenges. One of the most urgent problems is how to wandering nature of the nomads in a vast country like India
combine the retrieval results from different neighboring has had a toll on their children, who are denied even a
nodes into a single, integrated ranked list. In this paper, we semblance of formal or regular education. This results
propose a result merging algorithm to address the invariably in the children continuing with the traditional
challenge. Our algorithm is deterministic and does not nomadic culture. Maharashtra, a state in western India
require neighboring nodes to provide any information for shares borders with a number of neighbouring states. The
result merging, which makes it different from other nomadic tribes in this state, therefore, invariably wander
algorithms that require cooperation from participants. A amongst the neighbouring states. This paper is a report on
variety of experiments demonstrate that the new approach an exciting experiment with convergence of tradition and
is effective." ICT. The author and his team have studied the tribes in
detail and have evolved a methodology that involves the
3B-02.2 1568964111 use of all available technologies, based on relevance,
availability, portability, repeatability and flexibility. The initial
Quantitative Availability Prediction for a HA experiments carried out with three tribes in Maharashtra,
Messaging System have created the much needed motivation in the children
Xiaohu Yang, Wei Chen and mothers and given the assurance that the efforts will
bear fruit."
"How to predict the High Availability (HA) using quantitative
approach becomes important during system design phase. 3B-02.4 1568965051
Stochastic Reward Nets (SRN) is used to model the
behavior of the system. In this paper, we will adopt SRN as On Congestion Control in Multicast Networks
an analytic model to compute several interesting output Liansheng Tan, Qin Liu, Moshe Zukerman
measures that characterize availability features of a HA
Messaging System. This messaging service is the core part "We present a control-theoretic algorithm that regulates the
of a 24-hour non-stop safety-critical financial service. During source-rate at a multicast tree in a computer network. The
the system design phase, this prediction model is used to proposed control uses a simple controller and relies on
provide useful information for aiding special technology explicit rate feedback. We analyze the performance of the
choosing." proposed algorithm. Simulations are used to validate the
analysis and to demonstrate system stability, efficiency and
fast response."

Tencon 2005 Page 111

 3B-03
Session Title: Filter Design and Implementation 1
Chairperson: Amin Sadik
Day: Thursday
Date: 24th November 2005
Time: 9:30AM – 10:50AM
Room: P1

3B-03.1 1568963705 3B-03.3 1568964925

Electronically current mode all-pass filter with Efficient Structure for Single-Bit Digital Comb
only one grounded capacitor Filters and Resonators
Seangrawee Tongkulboriboon, Wijittr Petchakit, Amin Sadik, Zahir Hussain, Peter O'Shea
WiWat Kiranon A design technique for a single-bit digital comb filter is
A novel current mode first order all-pass filter with presented. The proposed filter response and performance
electronically tunable feature based on second generation are assessed in terms of signal-to-quantization-noise ratio
current controlled conveyor (CCCII) is presented. The (SQNR) and stability. It is found that the comb filter
configuration of the proposed circuit is simple and very possesses a distinct frequency response in broadband
attractive for monolithic circuits owing to using only one signal applications. The same technique is utilized to design
grounded capacitor and two CCCIIs. Phase shift and simulate a single-bit N-period digital resonator.
characteristic of the current mode all-pass filter can be Feedback loop filters can be used to tune the frequency
controlled by dc bias currents of the CCCIIs. The proposed response of the $\Sigma\Delta$ modulators. The proposed
circuit is verified by SPICE simulation and applied in current design technique is efficient in hardware implementation.
mode quadrature oscillator realization. The results from
simulations are included in order to confirm the validity of 3B-03.4 1568964952
the proposed circuits.
Design of DMB Nonuniform SIR-type BPF with
3B-03.2 1568964010 Embedded Tunable Pads Using LTCC

Identification of Nonlinear Differential Equation Byoung-hwan Lim, Sung Kyo Park, Hyang-guh-ri Ko,
Models from Generalized Frequency Response Ho-rim Chang, Chong-baek Park
Functions "Recently, RF systems have rapidly grown with the
extension of the mobile communication service. The mobile
Akshya Swain, Eduardo Mendes service companies are providing the satellite broadcasting.
A new algorithm called total least squares with structure Coinciding with current trend, the development of improved
selection (TLSSS) has been proposed to identify continuous satellite DMB (Digital Multimedia Broadcasting) tuner is
time differential equation models from complex frequency required. To improve the receiving sensitivity under the poor
response data. The algorithm combines the advantages of communication circumstance, it is necessary to design the
both the total least squares and orthogonal least squares BPF (Bandpass Filter) to transmit only desired signal
with structure selection (OLSSS). The error reduction ratio without distortion and loss. Besides high reliability, the
(ERR) feature of OLSSS are exploited to provide an miniaturization and lightweight are required for design of
effective way of detecting the correct model structure or mobile terminals. In this paper, we designed and fabricated
which terms to include into the model and the total least DMB nonuniform SIR-type BPF with embedded tunable
squares algorithm provides accurate estimates of the pads, which operates at 2642 §Ö and is embedded in the
parameters when the data is corrupted with noise. The substrate of RF module of tuner with LTCC. As a result, we
performance of the algorithm has been compared with the obtained the passband insertion loss of 2.4 dB and the
weighted complex orthogonal estimator and has been passband ripple of 0.08 dB. So this BPF is applicable to RF
shown to be superior. module of satellite DMB tuner."

Tencon 2005 Page 112

 3B-04
Session Title: Robust and Adaptive Control 1
Chairperson: Pubudu Pathirana
Day: Thursday
Date: 24th November 2005
Time: 9:30AM – 10:50AM
Room: P3

3B-04.1 1568964570 mechanism. Results of the experiment shows that bit-

stream implementation can offer a viable alternative for
Robust Nonlinear Coordinated Control of TCSC constructing controllers for systems with several
and Generator Excitation dimensions.
H. Lan, L.K Zhang
3B-04.4 1568965007
"An advanced robust nonlinear control scheme is proposed
for single-machine infinite-bus power system (SMIS) Robust terminal attitude navigation for
equipped with thyristor controlled series compensation autonomous vehicles
(TCSC). Firstly, for a class of multi-input multi-output
(MIMO) nonlinear system with disturbances, a disturbance
Pubudu Pathirana
attenuation method in the sense of gain is proposed "Here we define the terminal attitude of the pursuer with
through recursive method to construct storage function, and respect to the target and present a LQR and H1 control
then disturbance attenuation control law is obtained. approach to solving the problem of missile achieving a
Secondly, in view of the strong nonlinearity and uncertainty desired terminal attack/approach angle. The intercept or
of power system, a nonlinear model with disturbances is engagement criteria is defined in terms of both minimizing
constructed. Lastly, a new robust nonlinear coordinated the miss distance and controlling the pursuers body attitude
control law of TCSC and generator excitation is developed. with respect to the target at the terminal point. This
Simulation results on SMIS show clearly that the proposed approach in comparison to previous approaches consider
control law can attenuates the influence of external the relativistic approach of the missile with respect to the
disturbances on the system and transient stability of power target as opposed the absolute velocities of the two
system is improved." dynamic bodies, and have possible applications ranging
from autonomous vehicle entry in to a mother craft to nossle
3B-04.2 1568964643 engagements in on-flight refuelling or even in precision
missile guidance. Here we also suitably formulate the H1
Dynamic Modeling and Controller Designing for control ideas directly applicable for the underlying problem
Flexible Liquid-filled spacecraft and presents both state feedback and output feed back
Dongliang Peng results for the case of finite horizon and non-zero initial
conditions together with a optimal parameter value to
"The present paper deals with dynamics modeling and achieve a desired terminal characteristic in terms of the
controller designing for a large flexible spacecraft, such as original weighting parameters."
Tracking and Data Relay Satellite (TDRS) with several
flexible appendages and a liquid fuel tank. The basic 3B-04.5 1568965691
parameters for the equivalent mechanical model under low-
gravity are given when the pendulum motion of the tank is Efficient Utilization of Electrical Energy In
small enough such that it can be approximated linearly. Pumping Operations Based on Existing
Hybrid coordinate is used to describe the configuration of Conditions A Case Study
the large flexible liquid-filled spacecraft, and the D¯
Alembert principle in virtual work is employed to derive the P. Giridhar Kini, P. N. Sreedhar
dynamics equations. The equations obtained are simple "The intersection of the pump characteristics with the
and common, which can be applied to describe many system characteristics gives the design operating point for a
complicated open-chain spacecraft systems. A controller is given pumping system. Changes in conditions of fluid
designed using the variable structure control strategy. The viscosity, total head, throttle positions, supply system
control mission consists of maneuvering the large antennas parameters etc causes the operating point to shift on either
to track the pre-selected motion laws, stabilizing the side of the pump head-discharge curve, which affects not
spacecraft in an inertial space and suppressing the elastic only the hydraulic output of the pump, but also results in the
vibration of the appendages. A numerical example is given inefficient utilization of available electrical power. Use of
to prove the validity of the control laws." variable speed drives involves additional investment, which
if the pump is in operation for long hours every day
3B-04.3 1568964811 throughout the year with the number of such units in
operation being small in number, but is the best possible
Online Implementation of Servo Controllers option to achieve energy efficiency. But if the industrial
Using Bit-Streams utility has large number of pump units for various
Nitish Patel, Sing Kiong Nguang, George Coghill, processes, with individual operating times being quite
Akshya Swain insignificant, yet important in the process involved, variable
speed drives may not be the best possible solution. This
The present study proposes an alternate strategy for paper makes an attempt in presenting a simple method for
implementing controllers based on bit-streams. The key to making the pump to work at / or near its operating point
the success of bit-stream controllers is their ability to based on the exiting system conditions, with the least
represent continuous time signals using uniformly weighted possible changes to the working setup in addition to
binary streams. These can easily be manipulated using achieving the desired objective of efficient utilization of the
standard logic gates available on a FPGA and therefore available electrical power. The analysis is supported by a
offers significant advantages in real time implementation. case study."
The performance of the bit-stream controllers have been
demonstrated considering an example of a D.C. servo

Tencon 2005 Page 113

 3B-05
Session Title: Pattern Recognition 1
Chairperson: Kenichi Takahashi
Day: Thursday
Date: 24th November 2005
Time: 9:30AM – 10:50AM
Room: M1

3B-05.1 1568964823 3B-05.3 1568965116

Radial Tchebichef Invariants for Pattern Design Pattern Mining from Source Code for
Recognition Reverse Engineering
Ramakrishnan Mukundan Nilanjan Basu, Shubhashish Chatterjee, Nabendu
"Robust feature descriptors such as Zernike and Pseudo- Chaki
Zernike moments have been widely used in many pattern "Design patterns are micro architectures that have proved
recognition applications. However such moments require a to be reliable, easy-to implement and robust. There is a
coordinate transformation to the domain of the continuous need in science and industry for recognizing these patterns.
orthogonal functions followed by a discrete approximation of This paper aims toward development of a new method for
the integrals. This paper proposes a new class of invariants discovering design patterns in the source codes. The
based on discrete orthogonal polynomials which help to method provides a precise specification of how the patterns
eliminate the aforesaid problems. The mathematical work by describing basic structural information like
framework of radial Tchebichef moment invariants is inheritance, composition, aggregation and association, and
presented, and their feature representation capabilities are as an indispensable part, by defining call delegation, object
studied. The radial Tchebichef moments are constructed creation and operation overriding. We have tried to
using the discrete Tchebichef polynomials as the kernel, introduce a new XMLbased language, the Extensible
and they have a radial-polar form similar to that of Zernike Pattern Markup Language XPML, which provides an easy
moments. The discrete form of the moment transforms way for the users to modify pattern descriptions to suit their
make them particularly suitable for image processing needs, or even to define their own patterns or just classes
tasks." in certain relations they wish to find. The proposed method
is tested on four open-source systems, and is found to be
3B-05.2 1568964914 effective in discovering design pattern instances."
An On-line Recognition System for Handwritten
3B-05.4 1568965179
Pitman Shorthand
A Study on Face Similarity Using Knowledge
Graham Leeddham, Yang Ma, Colin Higgins, Swe
Processing Methods
Myo Hywe
"This paper investigates the potential of Pitman shorthand
Kenichi Takahashi, Hiroaki Ueda, Tetsuhiro
as a mean of fast real-time handwritten text entry Miyahara, Naoko Takeda
(>120wpm) on hand-held devices. An online system which "The human face plays an important role in social
combines input, recognition and transcription of Pitman interaction, such as communications, identifications, and
shorthand is proposed. The main recognition algorithms security. Thus, facial information processing becomes an
taken in the demonstration system are discussed and active subfield of the pattern recognition area. Facial
evaluated. Based on the implementation experience of the recognition/identification algorithms usually calculate the
system, future research directions are discussed." similarity between a face and faces stored in databases and
determine if they match or not. In those algorithms, the
measurement of similarity is important. This paper presents
knowledge about face similarity acquired by knowledge
information processing methods and discusses the results.
In the experiments, three kinds of questionnaires are used
to obtain training examples, and these examples are
processed by decision tree system C4.5 and inductive logic
programming system Progol."

Tencon 2005 Page 114

 3B-06
Session Title: Wireless and Telecommunication 1
Chairperson: Jianguo Huang
Day: Thursday
Date: 24th November 2005
Time: 9:30AM – 10:50AM
Room: M2

3B-06.1 1568964283 3B-06.4 1568965143

A Reconfigurable RF Circuit Architecture for Analysis of Right Angle Microstrip Slot Antenna
Dynamic Power Reduction Paitoon Rakluea
Daisuke Kawazoe, Hirotaka Sugawara, Takeshi Ito, "This paper presents analysis right angle microstrip slot
Masu Kazuya antenna for achive one frequency and dual frequency. The
“This paper proposes a reconfigurable RF circuit antennas consist of L-shaped slot and inverse L-shaped
architecture for dynamic power reduction. The architecture slot on the ground plane fed by microstrip line to obtain for
consists of RF circuits and a control circuit. The RF circuits one frequency. Dual frequency is operated by antenna
can be reconfigured by bias voltages of transistors and which L-shaped slot is integrated with inverse L-shaped slot
variable passive devices, and the RF circuit block can also on one ground plane. The characteristics of antennas are
be switched dynamically. Analog RF circuits usually have proposed and analyzed for instance input impedance, S11
redundant margin in circuit performance to compensate parameter VSWR and far field radiation patterns by using
PVT variations, and it also causes redundant power Finite-Difference Time Domain Method (FDTD). A
consumption. The reconfigurable RF circuit can reduce verification of our simulation procedure is confirmed by
power consumption by the dynamic reconfiguration, which comparing IE3D software which is based on the Method of
compensates the performance margin. We demonstrate Moments (MOM)."
dynamic power reduction for opamp and LNA."
3B-06.5 1568965215
3B-06.2 1568964766 ZWOS A Lightweight Operating System For
MAEC: A Movement-Assisted Energy Conserving Lightweight Embedded Wireless Devices
Method in Event Driven Wireless Sensor Jim Wu, Michael Melkonian
Networks
"A large number of commercial and public domain Real-
Ming Zhao Time Operating Systems (RTOS) are available to suit real-
"Energy is one of the most important resources in wireless time embedded applications. On the other hand, there are a
sensor networks. Recently, the mobility of base station has vast number of relatively simple devices, many based on 8-
been exploited to preserve the energy. But in event driven bit micro-controllers, which, due to limited resources, or lack
networks, the mobility issue is quite different from the of need, have no RTOS. We argue that the appearance of
continuous monitoring one because only a small portion of low-cost, lightweight wireless technologies, such as IEEE
sensor node has data to send at one time. The number of 802.15.4 (WPAN) and Zigbee, will lead to a demand for a
sensor node that forward traffic should be minimized to lightweight operating system that is easily portable, has a
prolong the network lifetime. In this paper, we propose a small footprint and will seamlessly integrate with protocol
movement-assisted energy conserving method which tries stacks, development tools and system software already
to reduce the amount of forwarding sensor node by provided by the wireless chip vendors. We propose a
directing the base station moving close to the hotspots. This generic zero-weight psuedo-RTOS, written entirely in C
method achieves good performance especially when language, which is suitable for this market. We call our
applied to a network with a set of cooperative mobile base system ZWOS for Zero Weight Operating System. We
station. Extensive simulation has been done to verify the demonstrate the ease of integration of ZWOS into a
effectiveness of the propose schema" development platform provided one of the major vendors of
WPAN and Zigbee chipsets."
3B-06.3 1568964990
Novel Dielectric Rectangular Slab Antenna With
Improved Performance
Samik Chakraborty, Bhaskar Gupta, Dipak Ranjan
Poddar
"Dielectric antennas have proved their utility in higher
frequency ranges due to low loss, light weight and high
performance. Rectangular slab is the commonest dielectric
antenna configuration used for diverse applications like
mobile and satellite communication, missile guidance etc.
Its directivity has been improved by introducing a truncated
hollow pyramid in front of the antenna which also resulted in
improvement of return loss and VSWR bandwidth
appreciably. Computer simulated results are presented for
the novel dielectric slab antenna with a pyramidal tapered
wave launching section fed by a rectangular metallic
waveguide."

Tencon 2005 Page 115

 3B-07
Session Title: Bioinformatics and Biomedical
Chairperson: Bertil Schmidt
Day: Thursday
Date: 24th November 2005
Time: 9:30AM – 10:50AM
Room: M3

3B-07.1 1568963568 3B-07.3 1568964545

Genetic Algorithm for Classification of Epilepsy Using Graphics Hardware to Accelerate
Risk Levels from EEG Signals Biological Sequence Analysis
R. Harikumar, R.Sukanesh, S.Raghavan Gerrit Voss, Andre Schroeder, Wolfgang Mueller-
"This paper introduces a Genetic Algorithm (GA) based Wittig, Bertil Schmidt
epilepsy risk level classifier from EEG signal parameters. "Sequence alignment is a common and often repeated task
The risk level of epilepsy is classified based on the in molecular biology. The need for speeding up this
extracted parameters like energy, variance, peaks, sharp treatment comes from the rapid growth rate of biological
and spike waves, duration, events and covariance obtained sequence databases. In this paper we present a new
from the EEG of the patient. A Binary Coded GA (BCGA) is approach to high performance biological sequence
then applied on the code converters classified risk levels to database scanning on graphics processing units. Using
obtain the optimized risk level that characterizes the patient. modern graphics processing units for high performance
The Performance Index (PI) and Quality Value (QV) and computing is facilitated by their enhanced programmability
receiver operating characteristics are calculated for this and motivated by their attractive price/performance ratio
method. A group of eight patients with known epilepsy and incredible growth in speed. To derive an efficient
findings are used for this study. High PI such as 92% for mapping onto this type of architecture, we have
BGA was obtained at a QV of 80%." reformulated the Smith-Waterman dynamic programming
algorithm in terms of computer graphics primitives. This
3B-07.2 1568964066 results in an implementation with significant runtime savings
on two standard off-the-shelf computer graphics cards. To
A Novel Statistical Cut-Strategy for DP-based our knowledge this is the first reported mapping of biological
Multiple Biosequence Alignment sequence alignment onto a graphics processing unit."
Cheng-Long Chuang
3B-07.4 1568964711
"In this paper, a novel cut-strategy is presented for solving
the problems of multiple biosequence alignment. Sequence Direct-Proportional Length-Based DNA
comparison is the most important primitive operation for Computing Implementation for Elevator
ana-lyzing of the bioinformatics data. The most fundamental Scheduling Problem
method for alignment of several biosequences is the
dynamic program-ming (DP) technique. The DP method is Mohd Saufee Muhammad, Zuwairie Ibrahim, Osamu
capable of finding opti-mal alignments for a set of Ono, Marzuki Khalid
sequences. However, when the length of the sequences "Implementation and experimental procedures to solve an
increased, the DP method is impracticable due to the elevator scheduling problem using direct-proportional
computational complexity is extremely high. Therefore, a length-based DNA computing approach are presented in
new method is proposed in this paper for reducing the this paper. All possible travel path combinations of the
computational cost of the DP technique. By recursively elevators are encoded by oligonucleotides of length directly
finding the structural features of the biosequences, the proportional to the elevators traveling time based on certain
proposed method can divide the biosequences into very initial conditions such as elevators present and destination
small alignment problem, which can be directly solved by floors, and hall calls from a floor. Parallel overlap assembly
DP, or other applicable methods that can pro-duce the is employed for initial pool generation and polymerase chain
results of alignment faster. By utilizing the ob-ject-oriented reaction for amplification. Gel electrophoresis is then
programming technique, the proposed method also performed to separate all the generated travel paths by its
provides low memory space consumption during execution. oligonucleotides lengths. The gel electrophoresis image is
Moreover, the proposed algorithm has been implemented in then captured to visualize the required optimal path.
an x86 demonstration program, and compares the effective Experimental result shows that DNA computing approach
and effi-cient performance with other known method." can be well-suited for solving such real-world problem of
this type of nature."

Tencon 2005 Page 116

 3B-08
Session Title: Sensors and their Applications
Chairperson: Jiankun Hu
Day: Thursday
Date: 24th November 2005
Time: 9:30AM – 10:50AM
Room: M4

3B-08.1 1568964160 3B-08.3 1568965417

H-Infinity Design for a Sonar-Platform Semi-Continueous Interpolation Algorithm for
Jiankun Hu Aspherical Surface Grinding
"In this paper, a high-precision control system design for a HyungTae Kim
real sonar-platform is reported. Specifically, how to transfer "This paper proposed a tracking algorithm for the tool path
a real engineering problem into a standard control design of an aspherical surface in grinding process. The concept of
problem is demonstrated. A novel, yet simple weighting the algorithm is that an operator puts in lens constants, the
function is proposed to incorporate the hard constraints and grinding tool follows the lens surface. The ideal equation of
system performance specifications. Sea wave disturbances, grinding tool path was derived from aspherical lens
water resistance due to ship navigating, non-linear friction in equation. The path was sectioned by arbitrary interval. The
cylinders, etc. are discussed. A new control structure with inputs of the alrogithm was determined from the the position
inner PID control loop is deployed. Simulation results error at the begin of the step, acceleration and velocity. The
demonstrate that this is a rather satisfactory design." velocity and the acceleration were updated by the position
error and velocity of the previous step. This made the tool
3B-08.2 1568965307 path smoother and increased continuity. The algorithm was
demonstrated on a 2-axis machine which had ball screw
Adaptive Self Configuration for Distributed Load and air guide. The tool moved on both of the convex and
in Sensor Networks concave surfaces by the proposed algorithm. The tendency
Muhammad Iqbal of the error was monitored by the change of the tool
diameter. The maximum position error was under submicro
"The dynamic architecture of the network is monitored by level."
exchanging so-called Network State Beacons (NSBs)
between key network nodes. The Beacon Exchange rate
and the network state define both the time and nature of a
proactive action to combat network performance
degradation at a time of crisis. It is thus essential to select
and update the beacon exchange rate (FX) according to the
variations in the load profile of the network. This paper
presents a novel localized method that for selecting and
updating the FX by adapting to the network load and energy
constraints. The results indicating that the model
reconfigures the network more effectively to achieve higher
throughput as well as greater network integrity, with minimal
resource overheads."

Tencon 2005 Page 117

 3B-09
Session Title: Speech Recognition and Extraction
Chairperson: Heyoung Lee
Day: Thursday
Date: 24th November 2005
Time: 9:30AM – 10:50AM
Room: M8

3B-09.1 1568964814 3B-09.3 1568964951

A Methodology for Improving PESQ Accuracy for Applying Hybrid Neural Network For Malay
Chinese Speech Syllables Speech Recognition
Fong Loong Chong, Ian McLoughlin, Krzysztof GOH KIA ENG
Pawlikowski We proposed a hybrid technique for speech recognition
"Unlike English and most other European languages, which applying 2 different neural network architecture. The
Mandarin Chinese has two unique characteristics, the proposed technique combines Self-Organizing Map (SOM)
consonant-vowel-consonant (CVC) phonetic structure and which known as unsupervised network and Multilayer
use of tones, which may affect its intelligibility after Perceptron (MLP) which known as supervised network for
processing by sound processing systems. Due to this, the Malay syllables speech recognition. We used a 2-
Perceptual Evaluation of Speech Quality (PESQ) objective dimensional self-organizing feature map as a feature
speech quality measurement system, which has been extractor which acts as a sequential mapping function in
proven effective in measuring the speech quality of sound order to transform the acoustic vector sequences of speech
processing systems processing English or some other signal into trajectories. The output of SOM is a matrix with
languages, may not accurately measure speech quality of same dimension and its elements take on binary values.
systems processing Chinese speech. An evaluation of The transformation of the feature vector simplifies the
PESQ was thus performed to investigate whether classification task by recognizer using Multilayer
intelligibility related problems that arise from the two Perceptron. The MLP classifies the binary trajectories that
characteristics are being considered in the computation of each syllable corresponds to. Experiments were conducted
speech quality. From the evaluation, it was revealed that on the 15 Malay syllables by 10 speakers for conventional
PESQ indeed does not consider them through low technique (MLP only) and proposed technique (SOM and
correlation between subjective intelligibility and PESQ MLP). Our technique has archieved better performance
scores. A method known as consonant amplification was where improves the accuracy up to 4.5%.
then proposed to improve correlation results for Chinese
speech. Its implementation in PESQ was investigated." 3B-09.4 1568965524
Sub-Nyquist Nonuniform Sampling and Perfect
3B-09.2 1568964818
Reconstruction of Speech Signals
On the Use of Forward Temporal Masking and
Heyoung Lee, Zeungnam Bien
Cumulative Distribution Mapping for Noisy
Speech Recognition "The discrete-time speech signals obtained by Nyquist
uniform sampler have two kinds of redundant samples. One
Eric Choi, Julien Epps originates from the time-varying instantaneous bandwidth.
"Robustness in the presence of various types and levels of The other originates from non-compactness of spectral
environmental noise remains an important issue for support in time-frequency domain. In this paper, the sub-
automatic speech recognition (ASR) systems. This paper Nyquist nonuniform sampling and the perfect reconstruction
describes a new noise-robust ASR front-end that employs a formula are proposed for removing the redundant samples
functional model of forward temporal masking combined due to the time-varying instantaneous bandwidth in speech
with cumulative distribution mapping based on MFCCs with signals. In the proposed method, the instantaneous
c0. Recognition experiments on the Aurora II connected sampling frequency varies, depending on the least upper
digits database reveal that the proposed front-end achieves boundary of spectral support of a speech signal in time-
an average digit recognition accuracy of 83.24% for a frequency domain. The definition of the instantaneous
model set trained from clean data and 90.32% for a model bandwidth, which determines the instantaneous sampling
set trained from data with multiple noise conditions. frequency and is used for generating the set of samples that
Compared with the ETSI standard Mel-cepstral front-end, represent continuous-time signals perfectly, is given."
the proposed front-end obtains a relative error reduction of
around 57% for the clean model set and 21% for the multi-
condition model set."

Tencon 2005 Page 118

 3B-10
Session Title: Watermarking
Chairperson: Tasso Athanasiadis
Day: Thursday
Date: 24th November 2005
Time: 9:30AM – 10:50AM
Room: M10

3B-10.1 1568964668 3B-10.3 1568965337

An intelligent lightweight Intrusion Detection Watermarking of Still Images in Wavelet Domain
System(IDS) based on Entropy Masking Model
Hu ZhengBing, V. P. Shirochin, Jun Su Bahareh Akhbari
"This paper presents an intelligent lightweight IDS, called "A new robust image adaptive digital watermarking
host-based forensic intrusion detection system,which is algorithm in wavelet transform domain is proposed in this
developed with forensics,biometrics,data mining and paper. The proposed method exploits Human Visual
watermark techniques.We use the forensics technique to System (HVS) characteristics and entropy masking concept
profile the user behavior in order to automate the to determine image adaptive thresholds for selection of
maintenance of user profile, data mining technique to find perceptually significant coefficients. The mark is embedded
out the cooperative attack, and watermark technique to in the coefficients of all subbands including the LL subband.
trace back the hackers or intruders. The goal of the system Experimental results show that the proposed method
is to detect the intrusion real-time, effectively and significantly improves watermarking performance over
efficiently." conventional methods, in the terms of invisibility and
robustness."
3B-10.2 1568964845
An Image Watermarking Scheme Based on
Orthogonal Moments
Pew-Thian Yap, Raveendran Paramesran
"This paper proposes a new image watermarking scheme
using the set of Tchebichef moments. The watermark is
constructed as an independent identically distributed (i.i.d)
Gaussian random vector that is imperceptibly inserted in a
spread-spectrum-like fashion into the perceptually
significant spectral components of the data. The watermark
is robust to distortion caused by JPEG coding, dithering and
image scaling. Experimental results are provided to support
these claims."

Tencon 2005 Page 119

 3C-01
Session Title: Convergence Issues
Chairperson: Suyong Eum
Day: Thursday
Date: 24th November 2005
Time: 11:20AM – 1:00PM
Room: M7

3C-01.1 1568964877 3C-01.3 1568964968

Performance Analysis of Adaptive Source Rate Sustainable Agriculture Mobile Commerce
Control Algorithm (ASRC) for VoIP (SamCom) Extending the rural grocery supply
Muhammad Usman, Noor Muhammad Sheikh chain through M-Commerce
"The quality of VoIP is highly degraded by network Nhiem Lu, Paula Swatman, Sylvie Perreau
dynamics like congestion of links, routing delays, packet "Despite the popularity of integrated grocery supply chain
loses etc. By changing the source-encoding rate adaptively management as a research area for both academic and
with network dynamics, a much better end-to-end quality of industry researchers, little work has yet been undertaken on
service can be achieved. This paper discusses some how to link primary producers into the chain. In this paper
already existing techniques for source rate control, we provide an introduction to SamCom (a joint university/
highlighting their limitations and presents a recursive industry project), which will investigate the feasibility of
algorithm which changes the encoding rate of voice using mobile phones to capture and transmit supply chain
transmitting source to achieve optimum QoS for VoIP under food product information from primary producers to the next
randomly varying network conditions. The algorithm is link in the supply chain supermarkets, restaurants, or export
tested for real time VoIP transmission and the results are partners. The project will focus on the Sustainable
compared with PCM Mu-law and G.728 fixed rate codecs." Agriculture sector and will comprise two sub-projects: an
investigation of the acceptability and use of M-Commerce
3C-01.2 1568965410 by primary producers themselves; and a technical
investigation of the techniques needed to capture and
Modeling of VoIP Conversational Speech in Noisy transmit food produce data using mobile phones from
Environment remote and rural areas. The paper provides an overall
Padungkrit Pragtong, Tapio Erke, Kazi Ahmed description of the project, briefly covering the existing
grocery supply chain literature, then describes the projects
"This paper proposes the Voice over IP (VoIP) traffic aims (generally, and in terms of the two sub-projects) and
models, which include the effect of noisy environment. The finally considers the potential contributions a successful
proposed models are based on the traffic data measured outcome might have to offer."
from the operating IP network of Telephone Organization of
Thailand (TOT) Corporation. It was found that the observed
distributions of talkspurt and silent durations considerably
differ from the standard model (ITU-T P.59). It is proposed
that a new state long burst representing the background
noise at the talkers place, is added into the model based on
continuous-time Markov process. The other two states,
normal speech and long silence, represent the normal
behavior of the VoIP user. If background noise occurs
during the speech, the model is classified as noisy speech
otherwise as noiseless speech. In the observed cases, the
data rate of noisy speech is about 64% bigger than the data
rate of noiseless speech."

Tencon 2005 Page 120

 3C-02
Session Title: Security Issues 2
Chairperson: Jiankun Hu
Day: Thursday
Date: 24th November 2005
Time: 11:20AM – 1:00PM
Room: P2

3C-02.1 1568964835 3C-02.4 1568965097

An Email Classification Scheme Based on An Information-Centric Approach to Data
Decision-Theoretic Rough Set Theory and Security in Organizations
Analysis of Email Security Atif Ahmad, A.B. Ruighaver
Wenqing Zhao, Yongli Zhu "Many organizations focus on a computing-centric approach
"The effects of spam on network is discussed. Unsolicited to information security whilst neglecting the security of
messages or spam, flood our email boxes, viruses, worms, information on paper and amongst personnel. This paper
and denial-of service attacks that cripple computer networks presents a model that is both media-independent and
may secret in spam. This threaten network security, stability information-centric, allowing organizations to pursue an
and reliability seriously. In this paper, A new scheme based integrated methodology towards analysing risks and
on decision-theoretic rough sets is introduced to classify providing information protection across all types of media.
emails into three categories -- spam, no-spam and Using this model to map information flows within business
suspicious. By comparing with popular classification and knowledge processes will quickly show that it will be
methods like Naive Bayes classification, Our anti-Spam almost impossible to control all risks, but the resulting
filter model reduce the error ratio that a non-spam is detailed risk profile may enable the organization to explore
discriminated to spam, and we can find potential security alternative processes with lower risks."
problems of some email systems."
3C-02.5 1568965105
3C-02.2 1568964617 The ARP Virtual Reality System in Addressing
Developing a Distributed Agent-based Computer Security Threats and Disaster Scenarios
Network Security Infrastructure Yang-Wai Chow, Ronald Pose, Matthew Regan
Jiankun Hu "Nations, corporations and political organizations around
"Most network security system tends to be centralized the world today are forced to deal with an increasing
which suffers a single point of failure and also incurs many number of security threats. As a result, various
overheads in exchanging messages. In this paper, we organizations must find ways to adequately equip and
develop a distributed agent-based intrusion detection prepare themselves to handle numerous dangerous and life
infrastructure for computer network security. Experimental threatening circumstances. Virtual reality is an extremely
results have validated the working scheme. Implementation important technology that can be used across a variety of
details are reported." different fields and for a number of diverse applications,
ranging from simulation training to visualization tools, in
3C-02.3 1568965075 order to prepare for and manage disaster situations. Head
Mounted Display (HMD) virtual reality systems attempt
Privacy Protection of Grid Services in a visually to immerse the user in a virtual environment.
Collaborative SOA Environment However, it is well recognized that latency, the delay in
responding to a users head movement, is a major
Debasish Jana, Amritava Chaudhuri, Datta Abhijit,
shortcoming that plagues immersive HMD virtual reality
Bijan Bihari Bhaumik systems. Excessive latency destroys the illusion of reality
"Grid computing has emerged as a powerful extension of that such systems attempt to present to the user. A
web for a request-response interaction between service hardware architecture known as the Address Recalculation
providers and service consumers. Grid computing caters to Pipeline (ARP) and a computer graphics rendering
this collaboration need of service providers and consumers technique called priority rendering, were designed to reduce
on business need within a service-on-demand style. While the end-to-end latency suffered by immersive HMD virtual
collating responses from multiple service providers, the reality systems. This paper discusses the benefits of using
master service provider or broker has to open individual the ARP virtual reality system in addressing security threats
sections of the form. But, the individual service provider is and disaster situations."
supposed to open the portion of the form designated for its
own filling up and not to intervene with anyone elses area.
This requires that the XML document be multi-parted and
have individual private key used for lock to undesired
service provider. Our work explores possibilities of
encrypting or signing a portion of a XML document so that
same XML document can float through multiple service
providers with different sections locked for view by a
particular server."

Tencon 2005 Page 121

 3C-03
Session Title: Filter Design and Implementation 2
Chairperson: Amin Sadik
Day: Thursday
Date: 24th November 2005
Time: 11:20AM – 1:00PM
Room: P1

3C-03.1 1568965033 3C-03.3 1568965073

A Steady-State Analysis of a CORDIC-Based Designing Low Cost Wideband Microstrip
Adaptive ARMA Lattice Filter Bandpass Filters
Shin'ichi Shiraishi, Miki Haseyama, Hideo Kitajima Cornelis Kikkert
"This paper analyzes steady-state properties of a CORDIC- "The design of RF filters where the bandwidth is much
based adaptive ARMA lattice filter. In our earlier work, greater than 10% of the centre frequency is difficult to
convergence properties of the filter in the non-steady state implement, since conventional filter design techniques such
have been clarified; however, its behavior in the steady- as those for coupled line or interdigital filters are applicable
state is not discussed. Therefore, we develop a distinct to filters with less than 10% bandwidth. This paper
analysis technique based on a Markov chain in order to describes a design technique for designing wideband RF
investigate steady-state properties of the filter. By using the filters and illustrates this with the design of a 5 resonator
proposed technique, the relation between step size and filter with a 50% bandwidth at a centre frequency of 1 GHz.
coefficient estimation error is revealed. We can use the The design technique produces filters with a remarkable
results of the analysis to obtain an efficient design of the accuracy, so that one-iteration filter designs can be
filter." produced reliably."

3C-03.2 1568965035 3C-03.4 1568965074

A Delayless Adaptive IFIR Filterbank Structure Unstable IIR-Like Filter Blocks for Generating
for Active Noise Control Linear-Phase FIR Filters: A Noise Analysis
Kasi Rajgopal, Venkataraman Sethuraman Peyman Arian
"Active noise control (ANC) applications involve adaptive "The authors have recently proposed two alternative
filters with hundreds of taps. These filters are associated techniques for generating narrow transition band linear-
with high computational burden depending on the number phase FIR filters. The implementation of both designs
of taps in the filter and may suffer from slow convergence if requires a set of demultiplexers and IIR filters, which
the reference signal spectrum has a large dynamic range. accounts for the efficiency of the designs. On the other
Subband and filterbank techniques have been proposed to hand, since the demultiplexers and the IIR filters go through
address these problems. Filter Bank Adaptive Filter (FBAF) an activation-deactivation cycle at different time slots, the
is one of the models, which can model a linear system noise analysis for the structure becomes extremely
exactly. The FBAF model is a parallel bank of filters with complex. This study aims to derive closed form formulae for
each bank (channel) comprising of cascade of a set of short the roundoff noise generated in the proposed structures."
length filter called interpolator followed by a sparse model
filter. Each channel consisting of the interpolator filter and 3C-03.5 1568965221
sparse model filter is identical to a classical IFIR filter,
hence the structure is called adaptive IFIR filterbank Stepped-Impedance Defected Ground Structure
structure. The filterbank basis functions are orthonormal, Nemai Karmakar, Rita Ling
which introduce delay in the signal path degrading the
performance of the adaptive filter. In this paper, we propose "The negative image of a microstrip stepped-impedance
a delayless adaptive filterbank structure based on IFIR lowpass filter (LPF) or hi-Z, low-Z filter is used on the
filterbank structure for active noise control. The proposed ground plane to produce wide and deep stopband. This
delayless adaptive filter has improved convergence rate structure is simpler in design and more compact in size
and shows better noise cancellation in the wideband and compared to conventional bandstop structures. The
narrowband ANC domain when compared with the simulation and measured results show good agreement in
conventional fullband adaptive structure. The computational 2.6 GHz 20-dB rejection bandwidth at 4 GHz center
complexity of the proposed structure is compared with the frequency. This distinct stopband characteristic can be
fullband structure." applied to devices such as filters, antennas and coplanar
waveguide to suppress harmonics and enhance
performances such as better gain and isolation."

Tencon 2005 Page 122

 3C-04
Session Title: Robust and Adaptive Control 2
Chairperson: Kieran Orchard
Day: Thursday
Date: 24th November 2005
Time: 11:20AM – 1:00PM
Room: P3

3C-04.1 1568965036 3C-04.4 1568965129

Sensorless Control Scheme for Permanent Robust Control Framework for Piezoelectric
Magnet Synchronous Motor Actuation Systems in Micro/Nano Manipulation
Prerna Gaur Hwee Choo Liaw, Denny Oetomo, Bijan Shirinzadeh,
"Sensorless control scheme for Permanent Magnet Gursel Alici
Synchronous Motor (PMSM) using state observer is “{Micro/nano manipulation has been identified as one of the
presented and this technique is then compared with sliding key enabling technologies for many emerging challenges.
mode controlled drive with sensors. Simulated results are Within this scope, piezoelectric actuators have played major
presented to substantiate the proposed estimation of the roles in achieving the required nano-resolution motion. This
speed. In PMSM based adjustable speed drives, speed paper proposes a robust control framework for piezoelectric
sensors are normally taken as its integral part, as they are actuation systems to follow specified motion trajectories.
used to sense the speed and provide the signal to enable The basic concept associated with this methodology lies in
the drive system to implement the control algorithm. the specification of a target performance and the robust
However, these sensors increase the size and cost of the control scheme formulation for piezoelectric actuation
motor, and also require regular maintenance. The use of systems to ensure the convergence of the position tracking
speed sensors is compromised on ruggedness and error to zero. This control methodology is attractive as its
reliability of the drive. Of the several schemes available for implementation requires only the knowledge of the
sensor less operation of PMSM State observer scheme is estimated system parameters and their corresponding
discussed in this paper: This scheme is compared with bounds, including bound of hysteresis and external
conventional sliding mode controller scheme of control of disturbances. Feasibility study of the framework for
PMSM." piezoelectric actuation systems in micro/nano manipulation
is described. Simulation results validated the suitability of
3C-04.3 1568965102 the proposed control approach."
A Resilient Guaranteed Cost Control Design Via
3C-04.5 1568965205
the Robust Two-Stage LQ Reliable Control
Microcontroller-Based Autonomous Light Craft
Chien-Shu Hsieh, Leewen Tyan, Sho-Shen Yang,
Chun-Chin Lin Liqiong Tang, Alex Wishart, Kieran Orchard
“This paper considers the LQR reliable control design "This Paper presents the development of a light craft
problem for uncertain discrete-time systems which are powered by a 1.3KW methanol engine and controlled by a
subject to parametric uncertainty in both state and input Cygnal 8051F020 microcontroller. The design of the light
matrices. A robust (fixed) version of the recently developed craft has a simple mechanical structure with easy control
ATSLQRC, named as the RTSLQRC, is proposed to solve features. Fuzzy logic is incorporated in the control system to
a resilient guaranteed cost control design problem for realize the intelligent control. The main characteristics of the
uncertain discrete-time systems in which the actuators may craft are discussed in the paper. In addition to the
have some faults. The connections of this new robust functionality, flight modes, mechanical and control aspects,
control with existing results are addressed. A numerical it also discusses the development of the control system
example is also given to illustrate the proposed method." based on a pair of microcontrollers"

Tencon 2005 Page 123

 3C-05
Session Title: Pattern Recognition 2
Chairperson: Saman Halgamuge
Day: Thursday
Date: 24th November 2005
Time: 11:20AM – 1:00PM
Room: M1

3C-05.1 1568965206 3C-05.4 1568965295

Attractor-Based Person Identification Document examiner feature extraction: thinned
Abbas Kouzani vs. skeletonised handwriting images
"An attractor-based person identification method is Vladimir Pervouchine, Graham Leeddham
presented in this paper. The self-similarity in images is "This paper describes two approaches to approximation of
exploited to develop the identification method. The method handwriting strokes for use in writer identification. One
is insensitive to partial image variations that are due to approach is based on a thinning method and produces
translation, rotation, or scale. The performance of the raster skeleton whereas the other approximates handwriting
system is evaluated by studying the results of its application strokes by cubic splines and produces a vector skeleton.
to a collection of images." The vector skeletonisation method is designed to preserve
the individual features that can distinguish one writer from
3C-05.2 1568965230 another. Extraction of structural character-level features of
handwriting is performed using both skeletonisation
Enhancing Episodic Associative Memory Models methods and the results are compared. Use of the vector
for Practical Use skeletonisation method resulted in lower error rate during
Leelani Wickramasinghe, Damminda Alahakoon the feature extraction stage. It also enabled to extract more
structural features and improved the accuracy of writer
"Associative memory models have been developed as identification from 78% to 98% in the experiment with 100
mathematical and cognitive models to represent human samples of grapheme ""th"" collected from 20 writers."
memory. The mathematical models have been successfully
applied in classification and pat-tern recognition tasks. But 3C-05.5 1568965306
there is a practical value in the cognitive representation of it
which has given rise to associative memory models known Using Particle Swarm Optimisation for Elastic
as Episodic Associative Memory (EAM). To date, the Bunch Graph Matching to Recognise Faces
practical application of EAMs is less common than the
mathematical models mainly due to representa-tion Rajinda Senaratne, Saman Halgamuge
deficiencies in EAMs. This paper proposes a novel "A new approach to face recognition, where Elastic Bunch
approach known as Episodic Associative Memory with a Graph Matching technique optimised with Particle Swarm
Neighborhood Effect (EAMwNE), which overcomes these Optimisation is proposed in this paper. It is a fully automatic
representative limitations and increases the practical value algorithm and can be used for databases where only one
of EAMs. EAM-wNE utilises the lateral relationships among image per person is available. A face is represented by a
fea-ture values, considering a neighborhood in achiev-ing face graph. A Face Bunch Graph is created as a
this. Consequently, each data pattern is repre-sented by a generalised representation of faces of various individuals.
number of memory traces (opposed to the single The face graph similarity between the Face Bunch Graph
representation in early models) depend-ing on the size of and the deformable graph that has to be fitted to the face in
neighborhood. In this paper, EAMwNE is experimented with the image is maximised by Particle Swarm Optimisation, to
classification tasks to demonstrate the enhanced locate the landmarks and thereby to find the optimal graph
capabilities of it. The accuracy of EAMwNE is further to represent a face. Improved results were obtained by this
enhanced by a reward mechanism." method as apposed to Elastic Bunch Graph Matching
without Optimisation. Results were further improved by
3C-05.3 1568965248 hybridising it with Principal Component Analysis."
Co-articulation Detection in Hand Gestures
Manas Bhuyan, Debashis Ghosh, Prabin Kumar
Bora
"The task of automatic gesture recognition is highly
challenging due to the presence of unpredictable and
ambiguous nongesture hand motions referred to as co-
articulation. Due to co-articulation that occurs during
transition from one gesture to the next, problem is
encountered in continuous hand gesture recognition. In this
paper, we have shown that co-articulation may be tackled
by identifying the key frames in the gesture video sequence.
We propose an acceleration feature for isolating the non
gestural motions between meaningful gestural indications
during trajectory guided recognition of hand gesture.
Experimental results demonstrate the effectiveness our
proposed scheme in identifying co-articulation in gestures
thereby improving the gesture recognition rate."

Tencon 2005 Page 124

 3C-06
Session Title: Wireless and Communication 2
Chairperson: Jianguo Huang
Day: Thursday
Date: 24th November 2005
Time: 11:20AM – 1:00PM
Room: M2

3C-06.1 1568965242 3C-06.4 1568965314

A Hybrid Method for the Calculation of Radiation High Gain Compact Circularly Polarized
Field from Opening in Shielding Enclosure Microstrip Antenna
Li Bin Nasimuddin Nasimuddin, Anand Verma, Karu
"This paper presents a new hybrid method for the Esselle
calculation of the radiation field outside the shielding A new combination of proximity coupled circular microstrip
enclosure. The method computes the tangential component antenna and a square quasi-planner surface mounted short
of the electric field on the opening by Finite Difference Time horn is proposed to increase the gain of a compact
Domain (FDTD) method to determine the equivalent circularly polarized parch by 4.2 dB. This antenna structure
magnetic current source. This magnetic current, once has gain 10.39 dBi and over 3 dB beamwidth ( 5.20) the
determined, can be used to ascertain the radiate field axial ratio (AR) is 0.11 dB. The AR bandwidth is 1.89%. The
anywhere external to the opening in a shielding enclosure. patch structure has 13% size reduction over normal circular
The accuracy of the proposed method is verified by microstrip patch antenna. Total thickness of stricture is only
comparing the exterior radiation fields with those calculated 0.22lemda.
by full FDTD and the traditional hybrid method. Simulation
results show that the new hybrid method significantly 3C-06.5 1568965353
reduces the computational effort and time, but still A Wideband Circularly Polarized Patch Antenna
maintains satisfactory results. It also shows that the
with Conical Radiation Pattern
proposed method is very effective for the far field in that the
computational effort, as compared with that of the near field, Ka Leung Lau
does not increase remarkably."
"A circularly polarized conical-beam patch antenna that
employs a hybrid feeding mechanism for wideband
3C-06.2 1568965262 operation is presented in this paper. It is implemented
60GHz Radios: Enabling Next-Generation through using a pair of L-probes and apertures, which are
Wireless Applications connected by the feeding network, to excite the circular
patch in orthogonal directions. The patch is supported by
James Howarth two plastic posts of 0.084£fo height in the air, where £fo is
"Up to 7 GHz of continuous bandwidth centred around 60 the free space wavelength at 2.53GHz. This antenna
GHz has been allocated worldwide for license free wireless possesses an impedance bandwidth of 28.2% (SWR < 1.5)
communications. Highly attenuated due to oxygen and an axial ratio bandwidth (AR < 3dB) of 28.3%. Also, it
absorption and small in wavelength, this band is ideal for has left hand circularly polarized (LHCP) conical patterns
extremely high data rate wireless data applications. These with the peaks located at around £c = +/-33degree over
include numerous WPAN/WLAN applications such as home these bandwidths. Moreover, its 3-dB gain bandwidth and
multimedia streaming. Traditional RF circuits used in this peak gain are 32.4% and 5.5dBi, respectively."
band are based on expensive compound semiconductor
technologies. However for viable consumer applications,
alternatives must be found. SiGe and CMOS based circuits
are showing promise for enabling this technology at a price
within reach of the consumer. This paper summarises a
joint project aimed at developing high rate consumer level
mm-wave wireless data systems. In particular, results to
date in our RF design efforts are summarised."

3C-06.3 1568965289
A Patch Loaded Dielectric Resonator Antenna
Ka Yan Hui, Kwai-Man Luk
A patch loaded dielectric resonator antenna (DRA) is
presented. It consists of a dielectric resonator surrounded
by a rectangular patch ring acting as a loading element
which produces a widen beamwidth in the antenna patterns.
A prototype antenna is designed and a 120o 3-dB
beamwidth is demonstrated. The measured radiation
patterns are stable across the operating frequency range.

Tencon 2005 Page 125

 3C-07
Session Title: Video Processing 1
Chairperson: Yung Lyul Lee
Day: Thursday
Date: 24th November 2005
Time: 11:20AM – 1:00PM
Room: M3

3C-07.1 1568963560 3C-07.4 1568964882

H.264 to MPEG-4 Resolution Reduction A Novel Error Detection and Copyright Protection
Transcoding Scheme for MPEG-2 Video
Jae-Ho Hur, Yung Lyul Lee Yuk Chung
"In this paper, we propose a heterogeneous trans- coding "This paper proposes an copyright protection and error
method of converting an H.264 video bitstream into an detection/recovery scheme for MPEG-2 video using Digital
MPEG-4 video bitstream. The proposed method reduces Watermarking and Error Control Coding (ECC) algorithms.
the spatial resolution for mobile terminals which support Instead of directly embedding into the MPEG2 video, the
only low resolution video bitstreams. When the H.264 video copyright information is first passed through an ECC
bitstream is transformed into the MPEG-4 video bitstream, encoder to generate an ECC coded watermark. The ECC
the conversions between the H.264 block types and the coded watermark is embedded into each 8x8 block of I
MPEG-4 block types are performed by minimizing the frame. At the decoder the watermark can be extracted and
distortion, and 4 4 block-based motion vector mapping is automatically corrected by ECC decoding. Besides, the
performed. The proposed transcoding method runs on embedded watermark information can also be used to
average 5.5 times faster than the cascaded transcoding detect and locate the transmission errors in the video
methods, for a degradation of the PSNR (peak-signal- to- stream. A video recovery method can then be applied to
noise ratio) of less than 0.5dB." improve the quality of the transmitted video. The
experimental results show that the proposed scheme is able
3C-07.2 1568963561 to detect errors in the video stream while decoding and
reconstruct copyright information more correctly than the
H.264 to MPEG-4 Transcoding Using Block Type traditional method."
Information
Jae-Ho Hur, Yung Lyul Lee 3C-07.5 1568965110
"In this paper, we propose a heterogeneous trans- coding A Scheduling Method for Real-time Decoding of
method of converting an H.264 video bitstream into an MPEG-4 Video Streams
MPEG-4 video bitstream. When the H.264 video bitstream Jinhwan Kim
is transformed into the MPEG-4 video bitstream, the
conversions between the H.264 block types and the MPEG- "Increasing use of bandwidth-hungry high-quality
4 block types are performed by minimizing the distortion, multimedia streams demands more and more computation
and 4 4 block-based motion vector mapping is performed. resources in the decoding stages at receivers. When poorly
The proposed transcoder runs 5.2 times faster than the scheduled, however, computation resources may become a
cascaded transcoding method, while maintaining the similar limiting factor to implementing seamless on-time playback
PSNR (peak-signal-to-noise ratio)." of multimedia streams. In this paper, we present a new
adaptive scheduling method for real-time MPEG-4 streams.
3C-07.3 1568963596 The main objective of our approach is to maximize the
number of concurrent multimedia streams without any loss
MPEG-4 to H.264 Transcoding in playback. The tasks of decoding MPEG-4 multimedia
Yung-Ki Lee, Yung Lyul Lee streams, considered to possess soft real-time
characteristics, are scheduled to use computation
"In this paper, a temporal resolution reduction transcoding resources based on their frame period and mean execution
method that transforms an MPEG-4 video bitstream into an time. The performance of the proposed technique is
H.264 video bitstream is proposed. The block modes and compared against those of similar techniques through
motion vectors in the MPEG-4 bitstream are utilized in the extensive simulation experiments. The simulation results
H.264 encoder for block mode conversion and motion reveal that the proposed method can effectively control the
vector interpolation methods. The proposed four types of use of computation bandwidth."
motion vector interpolation methods are developed not to
perform brute-force motion estimation again in the H.264. In
the experimental results, the proposed methods achieve
3~4 times improvement in computational complexity
compared to the cascade pixel-domain transcoding, while
the PSNR (peak signal to noise ratio) is degraded with
0.2~0.9dB depending on the bitrates."

Tencon 2005 Page 126

 3C-08
Session Title: Power Devices and Circuits 1
Chairperson: Yung C. Liang
Day: Thursday
Date: 24th November 2005
Time: 11:20AM – 1:00PM
Room: M4

3C-08.1 1568964167 3C-08.4 1568964815

Model of corona for transient study Choice of Reference Conductor in Three Phase
Mohammad Mohammadi, Behrooz Vahidi Systems A Paradigm Shift
This paper present a coron model on an HV transmission Harnaak Khalsa, Jingxin Zhang
line under the influence of AC and impulse voltage. The "The values of non-active (reactive) and total (apparent)
model is used in ATP in order to investigate the effect of power for three-phase systems change with choice of
corona on overvoltages. reference conductor. Presently, the neutral in a four-wire
system and virtual neutral in three wire systems is the
3C-08.2 1568964681 commonly used reference in definitions. In this paper, it is
shown that this is not necessarily the optimal choice of
Rapid Detection of Changes to Individual Modes reference. Thus, there is a need to define an optimal
in Multimodal Power Systems reference that has practical value. As a solution to this
Richard Wiltshire, Peter O'Shea, Gerard Ledwich problem, a new approach to determine the optimal
reference is proposed. The approach is simple but effective.
This paper describes a new method for rapidly detecting It allows extension of single-phase power definitions to 3-
modal deterioration in large interconnected multi-modal phase systems to produce unique values without the need
power systems. Any sudden detrimental change of an to address balanced and unbalanced cases separately. The
individual mode is detected using strategies derived from resulting definitions can be used to provide optimal non-
optimal detection theory. A statistical characterisation is active power compensation. The performance of this
used to establish reliable thresholds for detection of approach is illustrated with a case study. The case study
individual mode changes. uses three-phase three and four wire systems with different
resistive, inductive and capacitive load combinations. The
3C-08.3 1568964742 results of case study demonstrate the good performance of
Thermal Analysis of a Tubular Permanent Magnet the approach for the choice of reference conductor."
Linear Generator Using Multiphysics Solver
3C-08.5 1568964902
Siti Rohana Adi, Khalid Mohamed Nor, Wan Nor Liza
A Self-contained Direct Methanol Fuel Cell with
Wan Mahadi
Surface Tension Fuel Delivery
"This paper presents the thermal design aspect of linear
generator. A study is carried out to investigate the Yuming Yang, Yung C. Liang
temperature effects on the components in linear generator. "Direct methanol fuel cells (DMFCs) are now becoming the
One of the most important components in the linear leading candidate of next-generation power source
generator, Neodymium iron boron (NdFeB) permanent especially for portable electronics. However, so far almost
magnet is emphasized in this study since it provides a all DMFCs are fed with diluted methanol solution and
magnetic field source for the linear generator and the supplied by the active pump attached in the system. This
influence of its electromagnetic properties towards active component brings additional power consumption and
temperature will affect the performance of linear generator. lowers the entire system efficiency. In this paper, a simple
In this paper, the generator is coupled with the internal fuel delivery structure driven by surface tension is designed
combustion engine which tends to produce heat. Due to the and assembled for the DMFC system. Due to the different
high temperature from the combustion engine, and the heat surface tension of methanol and water, the methanol is
produced by the current carrying conductors, thermal expected to diffuse smoothly through the capillaries inside
analysis on different physical aspects has been done. In Teflon PTFE membrane into water side to form diluted fuel.
order to achieve an accurate result, the multiphysics solver The experimental results demonstrated that the DMFC kit
is used to analyze the thermal analysis for both electrical with the passive fuel delivery structure can achieve largely
and thermal problem. Finite element method (FEM) is used extended operating time period and higher fuel efficiency.
to observe this coupled field effect." Therefore, such passive fuel delivery component is believed
more applicable for micro scale DMFCs developing for
portable electronics."

Tencon 2005 Page 127

 3C-09
Session Title: Circuit Analysis and Design 1
Chairperson: Robert Brown
Day: Thursday
Date: 24th November 2005
Time: 11:20AM – 1:00PM
Room: M8

3C-09.1 1568964541 "A quasi-static equivalent circuit model of a dumbbell

shaped defected ground structure (DGS) is developed. The
"Analysis, Design and Implementation of an equivalent circuit model is derived from the equivalent
Active Clamp Forward Converter with inductance and capacitance developed due to the
Synchronous Rectifier" perturbed returned current path on the ground and the
narrow gap, respectively. The theory is validated against
Bor-Ren Lin
the commercial full-wave solver CST Microwave Studio.
"This paper presents the system analysis, circuit design and Finally, the calculated results are compared with the
implementation of an active clamp forward converter with measured results. Good agreement between the theory, the
synchronous rectifier. Compared with the conventional commercially available numerical analyses and the
forward converter, there is one auxiliary switch in the active experimental results validates the developed theoretical
clamp forward converter to recycle the energy stored in the model."
transformer leakage in order to minimize the spike voltage
at the transformer primary side. Therefore the voltage 3C-09.4 1568963897
stress of main switch can be reduced. The resonant circuit
based on the output capacitance and leakage inductance of Using Distributed Circuits to Model Power
the transformer will achieve zero voltage switching turn-on Capacitor Behavior
for both main and auxiliary switches to increase the circuit Robert Brown
efficiency. The synchronous rectifier is used at the
secondary side to further reduce the conduction losses. The "Distributed circuit modeling of stacked and spiral
operation principles of the active clamp forward converter capacitors is described. The modeling, executed on a
are analyzed in detail and the circuit performance is spreadsheet, allows what-if analysis of the effects of
compared with the conventional forward converter. The changes in capacitor specifications such as metal film
design procedure and example of active clamp forward resistivity and changes in geometric dimensions on
converter are presented. Finally experimental results are dissipation factor, equivalent lumped capacitance, film
presented for a converter with an ac input voltage of current density, film voltage and power distribution profile.
90~130Vrms, an output voltage of 5V/20A and operating at The system provides a complete range of data including
switching frequency of 150kHz are provided to verify the equivalent series resistance and complex impedance. The
zero voltage switching at turn-on." effects of varying degrees of edge disconnection are
provided in the model system. It provides a tool not only for
3C-09.2 1568965070 power circuit designers, but for manufacturers concerned
with design optimization and understanding of the effects of
Implementation of Bi-2223 HTS tapes as RF Coils varying degrees of in-use degradation. In this study, a
for 3T MRI System simple capacitor model with varying degrees of edge
disconnection is explored. The results correspond to
In-Tsang Lin, Chang-Wei Hsieh, Li-Wei Kuo, Wei-
common experience whereby degraded capacitors exhibit
Hao Chang, Herng-Er Horng, Hong-Chang Yang large increases in dissipation factor with little change in
"One way to reduce receiving coil noise in MRI scans is nominal capacitance. The results predict the existence of
using non-resistive high-temperature superconducting hot spots in a capacitor degraded by partial edge
(HTS) coils. They show advantages of much lower cost and disconnection."
easier fabrication over HTS thin film coils. In this work, we
built a 200mm in diameter Bi2Sr2Ca2Cu3Ox (Bi-2223) tape 3C-09.5 1568964832
HTS RF coil and demonstrated that the SNR of using the
HTS tape coil was 2.22 folds higher than that of the High-speed Weighing Using Impact on Load Cells
traditional copper coil for a phantom MR study. Test results Andrew Gilman, Donald Bailey
were in agreement with predictions, and the error of
predicted SNR gains and measured SNR gains is about "Many techniques employed in current weighing systems
0.9%. In vivo imaging, we also demonstrated that the SNR rely on the step response of one or a number of force
of using the HTS tape coil was 1.95 folds higher than that of transducers for making measurements. However, force
the traditional copper coil for a human wrist MR study. The transducers can have significant settling time, which limits
HTS coil can be expected to generate higher SNR gain the maximum rate at which measurements can be made.
after optimization. Further applications functional MRI is This paper presents a method of weight measurement that
under investigation to test the power of this HTSC system in utilizes the impact of an object on a load cell. The impact
our 3T system. Index Terms- MRI, Bi-2223, HTS tape coil, impulse can be approximated by the Dirac delta function
signal-to-noise ratio" and hence the output resembles the impulse response. The
size of the impact impulse and the weight of the object can
be found from this impulse response. Signal processing is
3C-09.3 1568965203
used to eliminate the long oscillatory response of the force
Quasi-Static Analysis of Defected Ground transducer, allowing the measurements to be made at a
Structure much higher rate."
Nemai Karmakar, Sushim Roy, Isaac Balbin,
Gerhard Swiegers

Tencon 2005 Page 128

 3C-10
Session Title: Power System Harmonics
Chairperson: Bor-Ren Lin
Day: Thursday
Date: 24th November 2005
Time: 11:20AM – 1:00PM
Room: M10

3C-10.1 1568963030 3C-10.3 1568963536

Three-Phase Capacitor-Clamped Active Power Series active power filter for current harmonic
Filter for Harmonic and Reactive Power and load voltage compensation
Compensation Bor-Ren Lin
Bor-Ren Lin "A multi-function of series active power filter is presented to
“This paper presents an active power filter based on a achieve current harmonic and load voltage regulation.
capacitor-clamped topology to compensate harmonics and Under the normal utility voltage condition, the series inverter
reactive power generated from the nonlinear loads. The is controlled to achieve current harmonic suppression so
balanced and sinusoidal line currents are drawn from the that the total harmonic distortion is reduced. If the abnormal
AC source under the balanced and unbalanced loads. The utility voltage is detected at the load side, the series inverter
capacitor-clamped topology is used in the voltage source is operated as a controlled voltage source to protect the
inverter to reduce the voltage stress of power sensitive loads against the voltage disturbances due to
semiconductors. A DC-link voltage controller, a hysteresis short-term abnormal voltage conditions. The reduced
comparator and a neutral point voltage compensator are switch-count inverter is used in the single-phase system to
used in the control scheme to maintain a constant DC bus compensate voltage disturbance for low power applications.
voltage of the inverter, track the compensated current and The front-end diode rectifier is used to supply the necessary
balance the neutral point voltage on DC side, respectively. active power under the voltage sag condition. The adopted
The experimental results are provided to verify the topology and control strategy are evaluated through several
effectiveness and validity of the proposed control scheme." simulation studies."

3C-10.2 1568963328 3C-10.4 1568964628

A New Approach for Harmonic Compensation Accurate Estimation of Harmonic Components of
Using Single- phase Hybrid Active Power filter Power Signals
Saad Mekhilef, Muhammad , Maliki Omar Akshya Swain, Nitish Patel, Lei Zhao
"This paper presents a new approach for harmonic control, "The present study proposes a novel technique of detecting
and its practical implementation for a single-phase hybrid and estimating the harmonic components of a signal under
active power filter. It is a combination of a series active the influence of significant noise. The proposed method
power filter and a shunt passive filter. The shunt passive combines the advantages of both singular value
filter is connected in parallel with a load and suppresses the decomposition (SVD) and orthogonal least squares (OLS)
harmonic current produced by the load, whereas the active algorithm. By introducing an error reduction ratio (ERR)
filter connected in series to a source acts as a harmonic test, the algorithm detects the correct frequency component
isolator between the source and load. For active filter present in the signal. The other parameters of the signal are
control, Sinusoidal Pulse Width Modulation (SPWM) is computed using SVD. The performance of the method has
developed and the modulation index is selected by been demonstrated considering a power signal at different
calculating the DC bus voltage of the active filter. The levels of noise and has been found to be satisfactory."
PSpice&#63194;, Matlab/Simulink&#63194; and MAX
PLUS II&#63194; softwares are used for simulation and 3C-10.5 1568965473
hardware implementation. ALTERA&#63194; FLEX 10K
controller and some peripheral circuits are used. Simulation
A High Efficiency Sub-Harmonic Self-Oscillating
and experimental results show that the proposed active Mixer for Modern Wireless Communication
power filter topology is capable of compensating the load System
current and voltage harmonic up-to the limit specified by Leung Chiu
IEC"
"A high-performance X-band sub-harmonic self-oscillating
mixer (SOM) is proposed. By using a re-configurable
transistor pair in conjunction with an open and a short circuit
stubs, the proposed SOM is capable of down converting
four RF frequency bands with center frequencies at 7.35,
8.65, 11.35 and 12.65 GHz with a conversion gain ranging
from 7.83 to 18.7 dB with very high power-added-efficiency.
The measured phase noise of the free-running oscillation is
V85.4 dBc/Hz at an offset frequency of 100-kHz. With
injection locking, the phase noise is reduced to less than
V100dBc/Hz at 10-kHz, without affecting the conversion
gains. This is the first demonstration of a multi-band sub-
harmonic SOM."

Tencon 2005 Page 129

 3D-01
Session Title: Mobile Cellular
Chairperson: Ahmet Sekercioglu
Day: Thursday
Date: 24th November 2005
Time: 2:00PM – 3:40PM
Room: M7

3D-01.1 1568964825 the mobility model. Through simulation, we show the

accuracy of our prediction algorithm and the simplicity of its
A Novel Joint Power and Rate Control Algorithm implementation."
in Cellular Radio Systems
Wenhu Xue 3D-01.4 1568965271
"The development of multimedia services in cellular radio Design and Performance Evaluation of an Inter-
systems requires effective control on the transmission rates System Handover Algorithm in UMTS/GSM
of users. In this paper, a novel joint power and rate control Networks
algorithm is proposed for multi-rate cellular radio systems. It
attains the maximum throughput with minimum transmission N. Saravanan, Nayaka Sreenivasulu, D. Jayaram, A.
powers. Simulations results show that the power Chockalingam
consumption performance of the proposed algorithm is "In this paper, we present a (UMTS <-> GSM) inter-system
better than selective power control (SPC) algorithm in handover algorithm which uses two pairs of UMTS and
feasible systems. In infeasible systems, both of its GSM HO thresholds (Ec/Io threshold for UMTS and RSSI
throughput and power consumption performance exceed threshold for GSM$); one pair of thresholds for making a
those of SPC." decision to handover from UMTS to GSM and another pair
of thresholds for making a decision to handover from GSM
3D-01.2 1568964950 to UMTS. We build a simulation model of a 37-cell
Measurement-based Performance Model of IP UMTS/GSM system consisting of an island of 7 UMTS cells
surrounded by 30 GSM cells. The simulation model
Traffic over 3G Networks includes mobility of all users in the system and spatial
Wee Lum Tan, On Ching Yue correlation of the log-normal shadow losses in the 2.1 GHz
(for UMTS) and 900 MHz (for GSM) frequency bands. We
"This paper presents a performance model of the
allow a test mobile to move along the boundary of the
throughput variations experienced by IP packets carried
UMTS and GSM cells and evaluate the handover rate
over 3G networks. The model parameters were obtained
performance as a function of the various handover
from end-to-end TCP measurements conducted in the
thresholds involved. We show that by properly building a
university campus served by different 3G networks.
hysteresis between the UMTS-to-GSM-HO threshold pair
Measurement tools were developed and deployed in a 3G
and the GSM-to-UMTS-HO threshold pair, a good
mobile phone and a server, allowing us to conduct TCP
compromise between call quality, and intra- and inter-
data transfers in various scenarios. One key observation is
system handover rates can be achieved.In otherwords, the
that time spent in state 0, corresponding to zero throughput
simulation results provide a set of design curves using
during handoff or rate switching, can last for several
which the UMTS and GSM HO threshold values for intra-
seconds. The results have been useful in not only
and inter-system handovers can be chosen judiciously."
understanding the performance of deployed 3G systems,
but also for validating our simulators for evaluating flow or
rate control algorithms for new IP-based applications."
3D-01.5 1568963777
On Hermitian Mapping for Direction-of-Arrival
3D-01.3 1568965006 Estimation of Coherent Cyclostationary Signals
Location based power control for Mobile devices Zhigang Liu
in a Cellular Network "By constructing the cyclic autocorrelation matrix with
Pubudu Pathirana Hermitian property, a Hermitian Mapping method(HMM) is
proposed which is appropriate for both cyclic MUSIC and
"This paper provides location estimation based power
conjugate cyclic MUSIC in this paper. By exploiting the
control strategy for cellular radio systems with location
cyclostationarity of the signals of interest(SOIs), this
based interference management scheme. Our approach
approach not only allows to select uncorrelated desired
considers the carrier -to-interference as a dependence on
signals ,but also has a better performance in the presence
the transmitter and receiver distance separation and
of multipath propagation.Theoretical analysis is presented
therefore an accurate estimation of the precise locations
and shows that the Hermitian Mapping method achieves a
can provide the power critical mobile user to control the
larger effective aperture of the array. Finally, simulation
transition power accordingly. In this fully distributed
results show the effectiveness of the proposed method."
algorithms, we propose using a Robust Extended Kalman
Filter (REKF) to derive an estimate of the mobile users
closest mobile base station from the users location, heading
and altitude. Our analysis demonstrates that this algorithm
can successfully track the mobile users with less system
complexity, as it requires measurements from only one or
two closest mobile base stations and hence enable the user
to transmit at the rate that is sufficient for the interference
management. Our power control algorithms based on this
estimation converges to the desired power trajectory.
Further, the technique is robust against system
uncertainties caused by the inherent deterministic nature of

Tencon 2005 Page 130

 3D-02
Session Title: Security Issues 3
Chairperson: Doan Hoang
Day: Thursday
Date: 24th November 2005
Time: 2:00PM – 3:40PM
Room: P2

3D-02.1 1568965109 3D-02.4 1568965259

Strategies for cost efficient security evaluations Investigating the relationship between Location-
Jussipekka Leiwo, Douglas Maskell Based Services and National Security
"Evaluating security products according to, for example, Holly Tootell
Common Criteria is costly and complex. In order to remain "Public awareness of national security has increased
competitive, organizations engaging in security evaluations significantly since September 11, 2001. Literature has
must aim at cost efficient evaluations. A criteria is discussed ways to respond or prevent breaches of National
established and used to assess the suitability of various security with some of these methods employing information
strategies for cost efficient security evaluations. The results technologies, including Radio Frequency Identification
suggest, not surprisingly, a dependency between strategic (RFID) and Global Positioning System (GPS). Other studies
role of security engineering and assurance and the have focused on the privacy impact the proposed solutions
suitability of cost efficiency strategies. While some will have. This research examines the technologies under
strategies may result in immature security engineering the umbrella of Location-Based Services (LBS). The
process, they may be most cost efficient if evaluations are preliminary findings of this research indicate that the
not in the core of the long term product engineering broader notion of LBS is what needs to be focused on in
strategies. On the other hand, strategic importance of order to understand the impact that they are having in the
assurance and security evaluations suggests that the costs effort to ensure national security."
of more repeatable security engineering process can be
justified by long term benefits." 3D-02.5 1568965164
3D-02.2 1568965112 A Secure Printing System for Ubiquitous Network
Security Models in the Password-Capability Yasushi Fuwa, Yuka Maruyama, Hiromine
System Kanazawa, Kenichirou Komatsu, Tomoaki Goshima,
Hisayoshi Kunimune
Dan Mossop, Ronald Pose
"This paper proposes a new secure printing system for
"Computer systems today are subject to the increasing networked environments. The system is designed to protect
problems associated with hackers, viruses, spyware and the data from being tapped, control the time of printing and
other undesirable activity. These problems often threaten pick up of the printed paper, control the user who receives
the confidentiality, integrity or availability of user data held the printed-paper, avoid leaks of information by making a
on the systems. Conventional systems such as Windows, mistake in the selection of the output printer, and support
Unix and Linux have not proven adequate protecting the copyright protection and a strictly limit the number of copies
data, in part due to their lack of support for strong security being printed. To accomplish its aims, the system encrypts
models. In this paper we discuss an operating system all communications and prints data. Also, it double encrypts
called the Password-Capability System. The system offers the key used for encrypting the print data by using the
a number of powerful security mechanisms. We show that public keys of the authorized printer to output the data and
these mechanisms allow the system, unmodified, to support the authorized user to receive the printed-paper.
the well-known Bell-LaPadula security model. Furthermore, Furthermore, it doesnft send any print data to the receiver,
we suggest minor modifications to the system to enable it to but only a key, and forces the receiver of the printed-paper
support a broad class of static and dynamic models, to be in front of the printer to starts decryption and output of
including Chinese Wall and Clark-Wilson." the print data. The system has been tested on-site showing
promising results."
3D-02.3 1568965207
Wireless Intrusion Detection: Not as easy as
traditional network intrusion detection
Zhiqi Tao, A.B. Ruighaver
"While wireless networks are growing in popularity,
monitoring these networks for abuse and intrusions is
almost non-existent. Although some Intrusion Prevention
Systems have recently appeared on the market, their
intrusion detection capabilities are limited. Real intrusion
detection in wireless networks is not a simple add on. This
paper discusses the fundamental difference between
wireless intrusion detection and traditional network intrusion
detection and identifies some of the challenges that will
need to be solved in designing and deploying a cost-
effective wireless intrusion detection system."

Tencon 2005 Page 131

 3D-03
Session Title: Filter Design and Implemetation 3
Chairperson: Amin Sadik
Day: Thursday
Date: 24th November 2005
Time: 2:00PM – 3:40PM
Room: P1

3D-03.1 1568965283 3D-03.3 1568965291

A Fast Transient Response Single Phase Active Active Filter Based On-Chip Step-Down DC-DC
Power Filter Switching Voltage Regulator
Mahesh Mishra, Linash P. K. C. H. Wu, Le-Ren Chang-Chien, Lih-Yih Chiou
"In this paper a fast dynamic response single phase active "Switching type step-down voltage regulators are commonly
power filter is proposed. The method proposed, used as the power source for the IC circuit loads. To ensure
compensates the non-linear load by improving the power that the voltage stability is well maintained, the switching
factor to the desired value and also eliminates the type voltage regulator usually uses LC low-pass filter in the
harmonics. The scheme is shown to work for both types output terminal. However, the inductor size and cost are two
compensation i.e. power factor correction plus harmonic of the major concerns in the circuit design of this kind of
elimination and harmonic elimination only under sinusoi-dal regulator. This paper presents an active-filter based step-
and non-sinusoidal supply voltage conditions. The active down dc-dc switching voltage regulator that could possibly
power filter (compensator) is realized using a sin-gle-phase make the whole circuit on chip. The analytical modeling and
voltage source inverter operated in a hysteresis current simulation results show that the proposed voltage regulator
control mode. The method is validated through digital can retain the same performance as the LC-based buck
simulation results using MATLAB. The results have been converter as we desired."
shown with ideal and practical compensator with an
emphasis on dynamic performance." 3D-03.4 1568965365

3D-03.2 1568965288 Applications of Low Voltage FGMOS based CCII

in Filtering
A Novel Three-Phase Active Power Filter Control
Algorithm with Unbalanced and Distorted Supply Susheel Sharma, Sher Rajput, Kirat Pal, Lalit
Voltages Mangotra, Sudhanshu Jamuar, Nor Noordin
"This paper aims at development of floating gate MOSFET
Mahesh Mishra, K. Karthikeyan (FGMOS) based low voltage, voltage-mode and current-
"In this paper, a novel control algorithm is proposed to mode circuits operating at ± 0.75 V supply voltages. Using
compensate the unbalance and non-linear loads in three- low voltage CCII, impedance converters have been
phase four-wire system using active power filter. The presented. The impedance converters have been used to
algorithm results in balanced and sinusoidal source currents simulate high value of inductance, which can be used for at
under balanced as well as unbalanced and distorted three- least 100 kHz. This circuit has also been used as
phase supply voltages. The algorithm makes the use of the capacitance multiplier. The circuit employs FGMOS based
positive sequence extraction of the supply voltage and the active elements besides grounded capacitors and is
theory of instantaneous symmetri-cal components. To suitable for chip implementation. Some new bandpass filter
illustrate the concept, a three-phase four-wire system with circuit has also been presented. Theoretical results have
unbalance and non-linear load is considered for been validated by PSpice simulations for 0.5 mm CMOS
compensation and detailed simulation results are technology with a supply voltage of ± 0.75 V."
presented"

Tencon 2005 Page 132

 3D-04
Session Title: Robust and Adaptive Control 3
Chairperson: Hidenori Nakazato
Day: Thursday
Date: 24th November 2005
Time: 2:00PM – 3:40PM
Room: P3

3D-04.1 1568965213 3D-04.4 1568965545

Timestamp based optimistic concurency control Magnetic Bearing Controller Tuning Utilizing
Quazi Mamun, Hidenori Nakazato Server/Client Technology
Optimistic Concurrency control demonstrates a few Esref Turker
improvements over pessimistic concurrency controls like "In the paper, an approach for self-tuning PD controller
two-phase locking protocol or time-stamp based protocol. parameters in magnetic bearings system was investigated
But the price of coarse detection of conflicts may and implemented on a Host PC (local PC). The task of
sometimes be high and consequently discounts the tuning PD controller parameters in a magnetic bearing
advantage of optimistic concurrency control protocol. In this system was extended to a remote PC. In this paper, remote
paper we have reduced the space of coarse detection of self-tuning was studied and implemented successfully.
conflicts of optimistic concurrency control protocol. Remote tuning of PD controller parameters was achieved
using Client/Sever Technology."
3D-04.2 1568965280
Greedy exploration policy of Q-learning based on 3D-04.5 1568965557
state balance Robust Stability Analysis of Delayed Neural
Yu Zheng, Siwei Luo, Jing Zhang, Zefan Zhang Networks with Polytopic Type Uncertainties
"Q-learning is one of the successfully established Yong He, Qing-Guo Wang, Wei Xing Zheng
algorithms for the reinforcement learning, which has been "We investigate the problem of global robust stability for
widely utilized to the intelligent control system, such as the delayed neural networks in this paper. We first utilize the
control of robot pose. However, curse of dimensionality and free-weighting matrices to express the relationship between
difficulty in convergence exist in Q-learning resulting from the terms in the system equation, and then apply the S-
random exploration policy and state-space quantization, so procedure to derive a stability condition for delayed neural
it is difficult for Q-learning to be applied to the control of networks. Next, we extend this result so as to establish a
complicated systems. In this paper, we propose a greedy global robust stability criterion for delayed neural networks
exploration policy of Q-learning with rule guidance. This with polytopic type uncertainties. Finally, we demonstrate
exploration policy can reduce the non-optimal action the usefulness of the obtained global robust stability
exploration as more as possible, and speed up the criterion and its improvement over the existing results by
convergence of Q-learning. Simulation results indicate the using a numerical example given in [21] for interval delayed
effectiveness of the proposed method." neural networks."

3D-04.3 1568965425
Real-Time Monitoring and Integrated Automation
Control in Steel Rolling Plant
Ruilin Pei, Lei Zhang, Li Wang, Ying Fang
"This paper mainly presents application of Human Machine
Interface (HMI) for real time tracing, monitoring, and
operation in steel manufacturing industrial automation
system, which are realized by the Programmable Logic
Controllers (PLC) to collect process data and transmitted by
the integrated computer networking--PROFIBUS DP. In the
tracing aspect, one of the newest monitor device-Process
Data Acquisition (PDA) implemented by ibaAnalyzer and
watchdog is applied in industry network, whose advanced
technology mainly include real time recording the data for
various of meters, motors, sensors and others, automatic
creating backup files for all recording data, and
automatically sending feedback to CPU and HMI operation
screen. All equipment can normally and successfully run by
using real time monitoring and integrated automation
control."

Tencon 2005 Page 133

 3D-05
Session Title: Pattern Recognition 3
Chairperson: Eustace Painkras
Day: Thursday
Date: 24th November 2005
Time: 2:00PM – 3:40PM
Room: M1

3D-05.1 1568965323 3D-05.3 1568965355

Face personalization: animated facial modeling A Framework for the Design and Implementation
approach using radial basis function of a Dynamic Face Tracking System
Edmond Prakash, Chen Chen Eustace Painkras, Charoensak Charayaphan
"In this paper, we propose a facial modeling system using “This paper presents a 3-tier Dynamic Face Tracking
our novel multilevel adaption procedure. We model a System (DFTS) framework based on Gabor Wavelets. The
template head model with physical structure to synthesize proposed system is composed of three major modules:
the various facial expression. To make the template model Face detection unit, Gabor decomposition & Similarity
suitable for different facial geometry, feature points computation unit and Face-lock refinement unit. First, we
compliant to MPEG-4 Facial Definition Parameters (FDP) present a neural network based face detection that is used
are labeled on this model. Then another personalized 3D for face localization in cluttered environments.
head with different geometrical information is labeled with a Subsequently, Gabor decomposition using a Gabor Wavelet
set of these feature points. An iterative multilevel training filter bank with 40 filters (8 orientations and 5 scales) of the
process is performed to get a transformation function from initial image with the detected face, is carried out for Facial
the template model to the personalized model. We use Feature Point (FFP) extraction and initialization. In the next
Compactly Supported Radial Basis Function (CSRBF) to stage, similarity algorithms along with Gabor Wavelet
assure the locality detail of the adapted model. A curvature decomposition are used to track the movements of FFPs in
based searching scheme is also used to find the feature the faces. Finally, a face-lock mechanism using a subset of
points in the iterative process." MPEG-4 Facial Definition Parameters (FDP) is utilized to
fine-tune the face tracking system under partial occlusion
3D-05.2 1568965350 conditions. Most of the current face tracking systems need
the FFPs to be labeled manually. They assume that the
3D Shape Extraction of Human Face in Presence targeted face has already been detected and localized. In
of Facial Hair: A Profilometric Approach our proposed approach, face detection followed by
Jay Yagnik, Kalpathi Ramakrishnan, Sai Siva, automatic initialization of FFPs using Gabor Wavelets is
done to facilitate face tracking in a seamless manner. Rapid
Kameswara Rao Lolla
prototyping of FPGAs to speed up the compute-intensive
“The increasing use of 3D modeling of Human Face in Face tasks was performed."
Recognition systems, User Interfaces, Graphics , Gaming
and the like has made it an area of active study. Majority of 3D-05.4 1568965413
the 3D sensors rely on color coded light projection for 3D
estimation. Such systems fail to generate any response in Skin-based Face Tracking Using Illumination
regions covered by Facial Hair (like beard, mustache), and Recognition
hence generate holes in the model which have to be filled Yuan-Pin Lin, Yi-Ping Chao, Chung-Chih Lin, Jyh-
manually later on. We propose the use of wavelet transform
based analysis to extract the 3D model of Human Faces
Horng Chen
from a sinusoidal white light fringe projected image. Our "Since skin-tone is luminance dependent, its sensitively
method requires only a single image as input. The method affected by illumination variation. Thus, the stability and
is robust to texture variations on the face due to space- accuracy of skin-based face tracker would decrease
frequency localization property of the wavelet transform. It dramatically while environment changed; especially
can generate models to pixel level refinement as the phase platform of the system is notebook or portable device. In
is estimated for each pixel by a continuous wavelet this study we propose an effective illumination recognition
transform. In cases of sparse Facial Hair, the shape technique, utilizing K-Nearest Neighbor classifier combined
distortions due to hairs can be filtered out , yielding an with adaptive skin models, to realize the real time tracking
estimate for the underlying face. We use a low-pass filtering system in various environments, which is more feasible
approach to estimate the face texture from the same image. than lighting compensation processing in real-time
We demonstrate the method on several Human Faces both implementation. We have demonstrated that the accuracy
with and without Facial Hairs. Unseen views of the face are of face detection based on the KNN classifier is higher than
generated by texture mapping on different rotations of the 90.05% in various environments. In real-time
obtained 3D structure. To the best of our knowledge, this is implementation, the system successfully tracks user face at
the first attempt to estimate 3D for Human Faces in 30 fps under standard notebook platforms. Although KNN
presence of Facial hair structures like beard and mustache classifier only initiates five environments at preliminary
without generating holes in those areas." stage, the system permits users to define and add their
favorite environments to KNN for computer access.
Eventually, based on this efficient tracking algorithm, the
low loading of CPU computation is benefit to other
intelligent applications of human-computer-interface."

Tencon 2005 Page 134

3D-05.5 1568965134
Human-Machine Interaction Compensating
Imperfection in Facial Expression Recognition
Mie Sato, Shang Yuyi, Masao KASUGA
“This study aims at comfortable communication between
humans and machines. We propose an interactive voice
output system for human-machine interaction. The facial
expression is paid attention to and automatically recognized
as one of the four expression candidates: ''joy,'' ''anger,''
''sadness,'' ''surprise.'' The facial expression recognition
uses features that correspond to expression-motion-
patterns. The 1st and 2nd expression candidates are
chosen from results of the facial expression recognition.
The information on these candidates is used for accessing
our voice database. The voice database contains voice data
corresponding to human feelings. In order to realize the
interactive voice output system, the voice data are
outputted with compensating imperfection in the facial
expression recognition, for a perfect facial expression
program has not been developed yet."

Tencon 2005 Page 135

 3D-06
Session Title: Wireless and Telecommunications 3
Chairperson: Donald Keogh
Day: Thursday
Date: 24th November 2005
Time: 2:00PM – 3:40PM
Room: M2

3D-06.1 1568965411 3D-06.4 1568961363

Spectrum Management of DSL Systems in Japan A Simplified Sphere Decoding Algorithm for
Shinichiro Yoshii MIMO Transmission System
"The Japanese broadband market is maturing due to Nor Noordin, Borhanuddin Mohd. Ali, Sudhanshy
expansion in the needs for richer IT lifestyles with higher bit Shekhar Jamuar
rate access. DSL technologies have played an important "In sphere decoding the choice of sphere radius is crucial to
role in such growth because of their lower cost and ease of excellent performance. In Chan-Lee sphere decoding based
deployment. In order to expand the coverage area and algorithm, the problem of choosing initial radius has been
compete with faster services such as FTTH, the DSL family solved by making the radius sufficiently large, thus
of technologies is diversifying, and there are various increasing the size of the search region. In this paper we
services on the market currently in Japan. It is very present maximum likelihood decoding using simplified
important to manage those different kinds of systems in sphere decoder as apposed to the original sphere decoder
order to minimize interference due to crosstalk. This paper for the detection of cubic structure quadrature amplitude
describes the Japanese spectrum management standard modulation symbols. This simple algorithm based on Chan-
put into effect this year." Lee sphere decoder allows the search for closest lattice
point in a reduced complexity manner compared to original
3D-06.2 1568965429 sphere decoder for multiple input multiple output system
Baseband Processor for A Global Positioning with perfect channel state information at the receiver.
Results show symbol error rate has stabilized even at very
System Receiver low initial value of the square radius. Keywords: sphere
Baijayanta Ray decoding, QAM, maximum likelihood detector and MIMO
system."
"In this paper, we propose a baseband processor for
obtaining the pseudo-range and navigation message in a
Global Positioning System receiver. The baseband
3D-06.5 1568965088
Processor so designed achieves highly reliable acquisition, The Capacity of Two Deterministic MIMO Radio
low hardware complexity and low power , at the cost of Channels
increased initial acquisition delay."
Donald Keogh
3D-06.3 1568965431 “This paper examines the channel capacity of a pair of
A Wideband Circularly Polarized L-Probe deterministic MIMO radio channels. Their characteristics
are dependent on a single parameter, the variation of which
Coupled Patch Antenna for Dual-Band Operation provides some insight into capacity multiplication or rather
Ka Leung Lau the loss of it."
"This paper presents the design and testing of a dual-band
circularly polarized patch antenna. The portion of the
antenna for upper band operation consists of two stacked
circular patches that are supported by the plastic posts in
the air. They are excited by four L-probes that are located
sequentially under the circumference of the lower patch. For
lower band operation, a larger circular patch is added below
the upper band patches. It is also coupled by four L-probes
sequentially. The proposed antenna has an impedance
bandwidth of 43.9% (SWR < 2) and a 3-dB axial ratio
bandwidth of 33% for the lower band, while an impedance
bandwidth of 55.2% (SWR < 2) and a 3-dB axial ratio
bandwidth of 44.7% can be obtained for the upper band.
Both bands have a peak gain of around 7.8dBi."

Tencon 2005 Page 136

 3D-07
Session Title: Video Processing 2
Chairperson: Min Cheol Hwang
Day: Thursday
Date: 24th November 2005
Time: 2:00PM – 3:40PM
Room: M3

3D-07.1 1568965253 3D-07.3 1568965343

Clustering based Video Transcoding for Fast Intra Prediction Mode Selection Scheme
Congestion Control in Wireless Environments Using Temporal Correlation in H.264
Srividya Gopalan, Balsree Ramanan, Amit Thawani, Min Cheol Hwang, Sung Jea Ko
Sridhar Varadarajan "H.264/AVC uses an intra prediction mode with 4x4 block
Multimedia applications such as video conferencing and and 16x16 block sizes for luma component and 8x8 block
media streaming require service guarantees and demand size for chroma component to improve the rate-distortion
strict timing requirements. Limited network resources along performance. However, the computational complexity of
with heterogeneous client devices usually act as a limiting H.264 encoder is dramatically increased due to the various
factor on the seamless global delivery of multimedia data. directional intra prediction mode. This intra prediction is
The key issue that needs to be addressed in adaptation and performed in every inter slice as well as every intra slice. In
delivery of such multimedia data to diverse devices is to this paper, we propose a novel fast H.264 intra prediction
strike a balance between usage of network resources and mode selection scheme that does not require extra
the quality of delivered content. In this paper we propose a computation. The proposed method uses the temporal
transcoding approach for video streaming to heterogeneous correlation between the intra predicted block in the current
mobile client devices over a wireless network with limited frame and the motion-compensated block in its reference
resources and bandwidth. Our approach is based on: (a) frames. This method can reduce the computational
Clustering client devices based on their capabilities to complexity while maintaining similar PSNR and bit rate. Our
determine the format of content suitable for them; (b) experimental results show the dominant performance of the
Adapting content appropriate to the determined content proposed algorithm."
formats by transcoding and achieving reduction in content
size by removing redundant information from content in the 3D-07.4 1568965464
process of transcoding; (d) Using subscriber distribution
and the devices they own to optimize the number of
Semantic Indexing of News Video Sequences: A
versions of content; and (e) Using transcoding to further Multimodal Hierarchical approach based on
reduce the size of content for each device cluster while Hidden Markov Model
maintaining the quality of content. Maheshkumar Kolekar, Somnath Sengupta
3D-07.2 1568965256 "In this paper, we propose a system that semantically
classifies news video at different layers of semantic
Design of Limited Buffer Size for Storing Multiple significance, using different elements of visual content. The
Reference Frames in a Video Codec classification hierarchy generates low-level concepts, and
concept hierarchy generates high-level concepts from low-
Chih-Chang Chen, Sheng-Shiung Chen, OScal T.-C. level concepts. Our classification hierarchy is based on a
Chen, Chien-Fa Chen, Fang-Chu Chen few popular audio and video content analysis techniques
"Multiple reference frames are recommended for motion like short-time energy and Zero Crossing Rate (ZCR) for
estimation and compensation to provide coding efficiency audio and Hidden Markov Model (HMM) based techniques
and error concealment in video codec. However, multiple for video, using color and motion as features. From these
reference frames consume more hardware cost for storing low level concepts of classification hierarchy, rule-based
themselves than one reference frame used in traditional classifier generates higher-level concepts. The proposed
video codec. Hence it is required to reduce the memory framework has been successfully tested with TV news
cells for storing multi-frames. In this work, we look closely at video sequences. Our results are very promising. We have
a macroblock (MB) of the currently decoded frame to used top-down hierarchical approach, which permits us to
determine the data type for storing itself, according to the avoid shot detection and clustering and consequently
coding parameters of blocks in this MB. In addition, we improves the classification performance. This has created
further estimate the computational complexity and memory- the great opportunity for supporting more powerful video
required distribution (CMD) of currently decoded frame from search engines."
the CMD of previous frame in a group of picture (GOP). An
optimized process is further performed to CMD of the 3D-07.5 1568965560
currently decoded frame to minimize the required
computational complexity with making a good use of the
Improved Algorithms for Implementation of
given buffer budget. Experimental results reveal that the MPEG2 AAC Decoder on FPGA
proposed scheme optimally constrains the memory usage Rajath Shenoy, Sudhir Naik, Sumam David
under a given buffer budget for storing five reference
frames as full as possible. In addition, because optimally "This paper discusses methods to implement the IMDCT
allocating the memory cells for storing the currently filter bank, Noiseless decoder, Inverse quantiser and Scale
decoded frame, a comparable video quality to that without factor application modules of MPEG-2 Advanced Audio
reducing memory cells is achieved." Coding decoder more efficiently when implemented on
FPGAs. The efficiency of the algorithms has been validated
through implementation on Xilinx Virtex II FPGAs."

Tencon 2005 Page 137

 3D-08
Session Title: Power Devices and Circuits 2
Chairperson: Sachin Pathak
Day: Thursday
Date: 24th November 2005
Time: 2:00PM – 3:40PM
Room: M4

3D-08.1 1568964992 33D-08.4 1568964701

Robust Position Control of a Synchronous Motor Three-phase fly-back AC/DC converter with novel
Using Discrete Sliding mode Model Following resonant snubber circuit
Controller Huang Xiao-jun, Gong Jian, Yang Xiao-lai
Phongsak Phakamach, Chumphol Akkaraphong "The technique of three-phase power factor correction in
"A Synchronous Motor position control system using a switch mode power supplies is still not mature and
Discrete Sliding mode Model Following Control or DSMFC currently, most of the schemes are based on the
is presented to achieve accurate tracking in the presence of combination of three single-phase power factor correction
load disturbance and plant parameter variations. The circuits with complex topology and control strategies
DSMFC algorithm uses the combination of model following resulting in high cost and low popularization. This paper
control and sliding mode control to improve the dynamics presents a three-phase fly-back AC/DC converter topology
response for command tracking. A design procedure is with a novel resonant snubber circuit that provides a high-
developed for determining the control function, the power-factor, low switching component stress , low cost and
coefficients of the switching plane and the integral control low complexity power supply interface to the utility such as
gain. The control function is derived to guarantee the mobile telecommunication base station. The circuit topology
existence of a sliding mode. A DSP-based synchronous of the converter is based on the three phase discontinuous
motor position control system using the DSMFC approach current mode (DCM) fly-back converter. This topology
is illustrated. Experimental results indicated that DSMFC makes the input current sinusoidal and in phase with the
system performance with respect to the sensitivity to input voltage, so the unity power factor (UPF) can be
parameter variations is greatly reduced. Also, its can obtained in theory. By means of a new resonant snubber
achieve a rather accurate tracking and is fairly robust to circuit which involves an assistant switching component, the
load disturbances." effective control of the peak current and voltage at the time
of the switching-off operations, nearly zero current switch-
3D-08.2 1568965500 off and lower switching loss is realized. In this topology, the
primary and the assistant switching components are
Leakage Current Estimation of H.V Insulators on operated separately, but both of them are controlled by
Power Distribution Lines constant duty ratio, so the control strategy of the system is
low complexity. The simulation results verify that the
Sachin Pathak
proposed converter achieves a near-unity power factor at
"Contamination-driven insulators failure causing wooden the input and a high quality DC voltage at the output side of
pole fire is a problem that increasingly disturbs H.V the interface."
distribution systems in Australia as well as other countries
of the world.. This paper presents an improved 3D-08.5 1568965028
mathematical model for leakage current with various wood
parameters and different environmental factors using PDSR in Phase Angle Space and SPM Based
dimensional analysis technique. The objective of the study Security Monitoring of Power Systems
is to understand the signature of the leakage current under Cun Dong, Yu Yixin
various environmental conditions such as humidity,
temperature and atmospheric pressure. The results "As the node power is controllable, the traditional dynamic
obtained from the proposed algorithm are compared to security region (DSR) is defined on power injection space.
experimental results of other researchers, and found to be The Phasor Measurement Unit (PMU) can measure and
in good agreement." transfer the state variables (phase angles) of power system
synchronously. How to make full use of the Synchronized
Phasor Measurement (SPM) in the security monitoring,
protection and control of power system is a challenging
research task. In this paper, we presented the concept of
practical dynamic security region (PDSR) in phase angle
space. A study on PDSR of power systems shows that
critical points satisfying transient stability in angle space can
still be described by one or a few critical hyper-planes. With
the PDSR in the node angle space, a concise expression to
calculate the transient stability margin of a current operation
point in each direction in node angle space is suggested.
Besides, this paper suggested an approach to acquire the
PDSR in low-dimensional angle space based on dynamic
equivalence of transient stability and a method to select
appropriate locations for partially equipping PMU, which
provide an innovative way of making use of the PMU
information to carry out dynamic security monitoring of
power systems."

Tencon 2005 Page 138

 3D-09
Session Title: Circuit Analysis and Design 2
Chairperson: Robert Brown
Day: Thursday
Date: 24th November 2005
Time: 2:00PM – 3:40PM
Room: M8

3D-09.1 1568965071 3D-09.4 1568965281

A New Variable Topology Genetic Coding for High speed and low area cost FPGA based
Evolutionary Hardware pattern matching
Chih-Yung Chen, Ching-Han CHEN Jian Huang, Yang Zongkai, Xu Du, Wei Liu
"Recently, the approaches that were all developed with "Intrusion detection and prevention system have to define
random circuit matrix as evolutionary framework have been more and more patterns to identify the diversification
widely used for evolutionary hardware and evolutionary intrusions. Pattern matching, the main part of almost every
hardware design. In this paper, a novel variable topology modern intrusion detection system, should provide
genetic coding including a new logic circuit representation exceptionally high performance and ability of
and corresponding evolutionary algorithm for evolutionary reconfiguration. FPGA based pattern matching sub-system
hardware design is proposed. In comparison with random becomes a popular solution for modern intrusion detection
circuit matrix based studies, the proposed approach is able system. But there is still significant space to improve the
to reduce searching space for evolutionary computation, FPGA resource efficiency. In this paper, we present a novel
obtain better flexibility for logic circuit encoding, and pattern matching implementation using the Half Byte
increase exploration possibility for more complex Comparators (HBC). HBC based pattern matching
evolutionary hardware. Simulation results demonstrate the approach can increase the area efficiency. But the
validity and performance of the proposed method." operating frequency will decrease. We also explored some
methods to improve the operating frequency in this paper.
3D-09.2 1568964124 The result shows for matching more than 22,000 characters
(All the rules in SNORT v2.0) our implementation achieving
A Survey of FPGAs for Acceleration of High an area efficiency of more than 3.15 matched characters
Performance Computing and their Application to per logic cell, achieving an operating frequency of about
Computational Molecular Biology 325 MHz (2.6Gbps) on a Virtex-II pro device. When using
quad parallelism to increase the matching throughput, the
Tirath Ramdas, Gregory Egan
area efficiency of a logic cell is decrease to 0.71 characters
"Molecular biocomputation workflows traditionally involve for a throughput of almost 8.5 Gbps."
days of compute time to align DNA/protein sequences.
Custom computing machines (CCMs) provide a means to 3D-09.5 1568965360
dramatically reduce alignment time, and FPGAs provide a
practical means to implement such CCMs. Software Logic Synthesis and Technology Mapping of
implementation of some sequence alignment algorithms MUX-based
suffer quadratic time performance, however CCM Maitrali Marik, Ajit Pal
implementations may be highly parallelized and
consequently provide linear time performance. Similarly, "The problem of logic synthesis and technology mapping for
CCMs may be used to accelerate workflows or operations Actel-1 MUX-based FPGAs in the context of low power
in a wide range of domains, however programming and applications has been addressed in this paper.
integration problems limit their usage, though progress has Decomposed BDD representation, which helps to deal with
been made to overcome these problems. With continued large Boolean functions, has been used in the technology
development of tools, devices, and integration solutions, independent as well as technology mapping phases of the
CCMs on FPGAs coupled to conventional systems present proposed synthesis approach. To minimize area, delay and
an effective architecture for high performance computing." power dissipation of the realized circuits, several novel
optimization techniques have been adopted in both the
3D-09.3 1568965218 phases. Efficacy of the proposed approach has been
demonstrated with the help of experimental results on a
Mapping of an N-Way Set Associative Cache to large number of ISCAS benchmark circuits."
An N-Square Grid for the LRU Algorithm
Mansour Jaragh
“Abstract: The CPU memory delay gap is considered
among the main issues in effecting the overall system
performance. The cache system is introduce in the system
to reduce the overall system delay time. The cache
replacement policy is one of the most important factors that
affect cache performance. With the trend of increasing
associativity in second-level caches, implementing an
effective replacement algorithm seems more challenging
than just eliminating conflict misses. The contribution of this
paper is in designing a practical circuit for the Least recently
used (LRU) algorithm using a square grid that is easily
mapped to an nxn grid of memory elements. The suggested
method is seen to be mapped in a one to one basis to its
corresponding hardware."

Tencon 2005 Page 139

 3D-10
Session Title: Machine Modelling and Simulation
Chairperson: Tasso Athanasiadis
Day: Thursday
Date: 24th November 2005
Time: 2:00PM – 3:40PM
Room: M10

3D-10.1 1568963930 the data obtained using nano indenter with that obtained by
using new technique was briefly discussed. It has been
A Novel Half-Way Shifting Bezier Curve Model shown that the measured nano hardness values of the
Ferdous Sohel, Laurence Dooley, Gour Karmakar material sensitively depend on the applied load of
indentation. It is also shown that the measured nano
"Bezier curves can cause a considerable gap to occur hardness is smaller than nano indenter hardness.
between the approximation curve and its control polygon,
due to considering only the global information of the control 3D-10.4 1568965015
points. In order to reduce this error in curve representations,
localised information needs to be incorporated, with the Energy Management Program for Use of
main philosophy to narrow down the gap by shifting the Induction Motors Based on Efficiency Prediction
Bezier curve points closer to the control polygon. To
integrate this idea into the theoretical framework of the Thanaphiwat Phumiphak, Thanapol Kedsoi, Chaiwut
classical Bezier curve model, this paper presents a novel Chat-uthai
Half-way shifting Bezier Curve (HBC) model, which "Electric drive systems are currently responsible for the
automatically incorporates localised information along with largest part of the electricity consumption. As the cost of
the global Bezier information. Both subjective and objective energy increases, it is increasingly more important to
performance evaluations of the HBC model using upon a remove the inefficient induction motors from operation. This
number of objects having arbitrary shape confirm its paper proposes an energy management program that can
considerable improvement over the classical Bezier curve help the plants to determine how much saving will be
model without increasing the order of computational achieved by using more efficient motors. This paper
complexity." focuses on the use of motor field efficiency estimation of
motors without the removal of motor from service and
3D-10.2 1568964850 without the output power measurement, and the use of
The Influence of IGV and Relative Humidity of Air motor efficiency prediction based on the manufacturer
information. The motor efficiency evaluations are obtained
on Performance Map of Axial Flow Compressor from the equivalent circuit method. The test results indicate
Zhang Xuelei that this program is tool of accurate economic analysis for
the purchase scenario. As result, it is possible to conduct
"Compressors constitute an integral part of most turbo
on-site energy audits of existing motors in order to
machinery systems and are widely used in many chemical,
determine the potential savings and payback period with
industrial, and aerospace applications. The performance
great confidence."
map of compressor is influenced by IGV closure angle and
relative humidity of air. The calculation model of axial flow
compressor is built in the paper, and the influence of IGV
3D-10.5 1568965346
and relative humidity on performance map is analyzed in Application of Power Source Element in Power
the paper." Factor Correction
3D-10.3 1568964083 Kamon Jirasereeamornkul, Marian Kazimierczuk,
Itsda Boonyaroonate, Kosin Chamnongthai
New method for nano hardness measurement of
single crystal Aluminum "A new application of power source element in power factor
correction is proposed. The power source is inserted
Shahjada Pahlovy between a bridge rectifier diodes and a large capacitance
Nano hardness is an important mechanical properties of bulk capacitor to increase the conduction angle of the
materials most used to characterize the wear resistance of rectifier diodes. The power from the DC bus is transferred to
materials. In this present paper a procedure is developed power source via a loss free resistor connected with the DC
for calculation of nano hardness by nano indentation and an bus. Analysis, design example, control scheme, and
Atomic Force Microscope (A.F.M).A three sided pyramidal simulation results are given in this paper."
diamond tip has been used for the indentation experiment.
Experiment has been concluded on single crystal Aluminum
(Al). The load displacement data have been used to obtain
the nanoindentation hardness and the room temperature
nano hardness properties of single crystal Al have been
investigated at different indentation loads. Comparison of

Tencon 2005 Page 140

Name and Initial Session Paper Title

Abbaspour, A. 1C-10.3 An Analytical Method for the Reliability Evaluation of Wind Energy Systems

Abdel-Akher, M. 2B-12.9 Development of Power System Analysis Software Using Object Components

Abdul Jabar, S. 2B-02.4 Alternative Architectures for Bidirectional Single Mode Fiber SuperPON 512 ONU, 100 km

Abdul Kareem, S. 2C-02.1 Ontology-Based Multimedia Presentation Generation

Abhijit, D. 1D-10.3 Dynamic User Credential Management in Grid Environment

Abhijit, D. 3C-02.3 Privacy Protection of Grid Services in a Collaborative SOA Environment

Addie, R. 2D-01.2 Comparing different approaches to the use of DiffServ on the Internet

Adzir Mahdi, M. 1B-07.5 Threshold Features of L-Band Linear Cavity Multiwavelength Brillouin-Erbium Fiber Laser

Double-Pass Discrete Raman Amplifier with FBG for Pump Power Optimization and Improved
Adzir Mahdi, M. 2C-12.3
Performance

Ahmad Ansari, S. 1D-07.3 Development of an Analog Electronic Wattmeter

Ahmad, A. 3C-02.4 An Information-Centric Approach to Data Security in Organizations

Double-Pass Discrete Raman Amplifier with FBG for Pump Power Optimization and Improved
Ahmad, A. 2C-12.3
Performance

Ahmad, J. 2D-12.10 Target Tracking in an Image Sequence Using Wavelet Features and a Neural Network

Ahmad, N. 1B-09.1 Optimization of UTRA SHO Algorithm

Ahmad, N. 1B-09.4 Downlink Soft Handover Mechanism Performance in WCDMA System

Ahmad, S. 1D-07.2 Frequency and Flexibility Based Cell Phone Keypad Layout

Ahmad, S. 1D-07.3 Development of an Analog Electronic Wattmeter

Ahmed, J. 2D-12.10 Target Tracking in an Image Sequence Using Wavelet Features and a Neural Network

Ahmed, K. 3C-01.2 Modeling of VoIP Conversational Speech in Noisy Environment

Akhbari, B. 3B-10.3 Watermarking of Still Images in Wavelet Domain based on Entropy Masking Model

Akhlaghian Tab, F. 2D-06.2 Scalable multiresolution Image Segmentation and Its Application in Video Object Extraction Algorithm

Robust Position Control of a Synchronous Motor Using Discrete Sliding mode Model Following
Akkaraphong, C. 3D-08.1
Controller

Alahakoon, D. 3C-05.2 Enhancing Episodic Associative Memory Models for Practical Use

Al-Araji, S. 2C-08.3 Extended Locking Range DTL using Chaos Technique

Alarcon, L. 1B-07.1 Spiral Inductor Coupling on 0.25 um Epitaxial CMOS Process

Alarcon, L. 1D-11.1 Design Methodology for CMOS Low-Noise Amplifiers Using Power Matching Techniques

Alarcon, L. 1D-07.4 Monolithic Spiral Inductors for a 0.25 um Digital CMOS Process

Al-Dabbagh, M. 2B-12.10 Applications of FACTS Controller for Improving Power Transmission Capability

Alici, G. 3C-04.3 Robust Control Framework for Piezoelectric Actuation Systems in Micro/Nano Manipulation

Al-Qahtani, F. 1C-12.4 A Space-Time Block-Coded PCC-OFDM Scheme for Flat Fading Wireless Channels

Recovery from Segmentation Failures Using Photometric Invariance in an Interactive Object

Altab Hossain, M. 1D-03.2
Recognition System

Ananthakrishnan, M. 3B-02.3 Educating Nomadic Children: An Experiment with the Convergence of Technologies

Anantrasirichai, N. 2D-03.2 Designing Rectangular Slot Loop Antenna for WLAN Application

Tencon 2005 Page 141

Name and Initial Session Paper Title

Anderson, T. 2B-02.1 Optical Monitoring Issues in Highly Transparent Optical Networks

Andrew, L. 2D-03.1 Evaluation of FAST TCP in Low-Speed DOCSIS-based Access Network

Andrew, L. 2D-03.3 Price-based Max-Min Fair Rate Allocation in Wireless Multi-hop Networks

Angelo Festin, C. 1D-02.5 An Enhanced Framing Strategy for Jitter Management

Annaiyan, U. 1B-07.2 Development of a conductive photoresist with a mixture of SU-8 and HCl doped polyaniline

An Adaptive High Resolution Measuring System for Characterizing Transformers Subjected to

Anniyappan, P. 1D-06.5
Broadband Excitation

Arai, T. 1C-08.1 FES as Biofeedback for an EMG Controlled Prosthetic Hand

Arai, T. 1C-09.3 Use of Turbo Coding for Embedded Transmission Scheme in Optical CDMA

Arian, P. 3C-03.4 Unstable IIR-Like Filter Blocks for Generating Linear-Phase FIR Filters: A Noise Analysis

Armitage, G. 1C-02.2 Evaluating the impact of DNS and HTTP session characteristics on consumer ISP web traffic

Armitage, G. 2B-11.3 Validation of the Random Waypoint Mobility Model Through a Real World Mobility Trace

Armitage, G. 2B-05.4 End-users resource consumption of spam and a 3D anti-spam evaluation framework

Armitage, G. 2C-11.7 Transparent IP layer Interception on Enterprise 802.11b/g Networks

Armitage, G. 2C-11.1 An Architecture for Automated Network Control of QoS over Consumer Broadband Links

Armstrong, J. 1B-12.5 Error Performance Analysis of Clipped Alamouti Space-Time Coded OFDM Systems

Armstrong, J. 1C-12.1 Alamouti Coded OFDM in Rayleigh Fast Fading Channels - Receiver Performance Analysis

Asuhaimi Mohd Zin, A. 2B-06.2 COMTRADE-Based Fault Information System for TNB Substations

Transmission of Compressed Multimedia Data over Wireless Channels Using Space-time OFDM with
Athanasiadis, T. 2B-09.5
Adaptive Beamforming

Athaudage, C. 2C-09.2 Low-Complexity Diversity Enhanced OFDM for Wireless Communications

Atkinson, I. 1D-04.2 High Speed Over Ocean Radio Link to Great Barrier Reef

Atov, I. 2D-03.1 Evaluation of FAST TCP in Low-Speed DOCSIS-based Access Network

Attakitmongcol, K. 2C-06.2 Power Quality Disturbance Recognition Using Wavelet-Based Neural Networks

Azad, M. 1D-07.2 Frequency and Flexibility Based Cell Phone Keypad Layout

Aziz, S. 2C-05.4 A Bus Level SystemC Model for Evaluation of Avionics Mission System Data Bus

B. H. Tay, D. 1D-08.1 Optimizing the FBI Fingerprint Compression Algorithm for BLACKFIN Processor

A Novel Resource Advertisement and Discovery Model for Ubiquitous Computing Environments using
Bagheri, E. 2D-08.4
Mobile Agents

Bailey, D. 2D-11.1 Design Patterns for Image Processing Algorithm Development on FPGAs

Bailey, D. 2B-04.1 Bias Characteristics of Bilinear Interpolation Based Registration

Bailey, D. 3C-09.5 High-speed Weighing Using Impact on Load Cells

The Development of a Novel Image Quality Metric and a Synthetic Colour Test Image for Objective
Bailey, D. 1C-03.2
Quality Assessment of Digital Codecs

Balbin, I. 3C-09.3 Quasi-Static Analysis of Defected Ground Structure

Bansal, B. 2D-11.8 Creation of a Grid environment and Grid Portals

Bapat, A. 2C-08.1 An Asynchronous BI-Polar Ternary ADC

Tencon 2005 Page 142

Name and Initial Session Paper Title

Basu, N. 3B-05.3 Design Pattern Mining from Source Code for Reverse Engineering

Bawa, S. 2D-11.8 Creation of a Grid environment and Grid Portals

Beckett, P. 2C-05.5 A Nanowire Array for Reconfigurable Computing

A Game Theoretic Framework for Call Admission Control in CDMA Mobile Multimedia Systems with
Benjapolakul, W. 1B-01.2
Traffic Asymmetry between Uplink and Downlink

Besar, R. 1C-08.2 Bluetooth-enabled ECG Monitoring System

Bhattacharya, S. 1B-04.1 Automated Translation of Behavioral Models using OCL and XML

Bhattacharya, S. 2D-08.1 User Model to Design Adaptable Interfaces for Motor-Impaired Users

Bhuyan, M. 3C-05.3 Co-articulation Detection in Hand Gestures

Bhuyan, M. 1D-03.5 Designing of Human Computer Interactive Platform for Robotic Applications

Biansoongnern, S. 1D-06.1 Minimization of transmission loss with consideration of transformer taps constraint

Bickerdike, P. 1C-01.3 Hybrid Delay/Blocking Wireless Systems with Greedy Resource Allocation

Bien, Z. 3B-09.4 Sub-Nyquist Nonuniform Sampling and Perfect Reconstruction of Speech Signals

Bihari Bhaumik, B. 1D-10.3 Dynamic User Credential Management in Grid Environment

Bihari Bhaumik, B. 3C-02.3 Privacy Protection of Grid Services in a Collaborative SOA Environment

Bin, L. 3C-06.1 A Hybrid Method for the Calculation of Radiation Field from Opening in Shielding Enclosure

Geometrical and Phasor Representation of Multi-level Amplitude-Phase Modulation Formats and

Binh, L. 1B-05.5
Photonic Transmitter Structures
Long-haul ASK and DPSK Optical Fibre Transmission Systems: Simulink Modeling and Experimental
Binh, L. 1C-05.2
Demonstration Test-Beds

Bishop, A. 3B-01.3 Sensor Fusion Based Localization of a Mobile User in a Wireless Network

Biswas, P. 2D-08.1 User Model to Design Adaptable Interfaces for Motor-Impaired Users

Bo, Y. 1D-01.4 A Traitor Tracing Scheme Based on the RSA Scheme

Boonyaroonate, I. 3D-10.5 Application of Power Source Element in Power Factor Correction

Bose, S. 2B-11.7 SERA: Service Ratio Based Preemption for Minimizing Throughput Loss in Networks

Bose, S. 1B-02.1 CLA-QOS: A Cross-Layer QoS Provisioning Approach for Mobile Ad-hoc Networks

Bose, S. 1B-02.2 Modifying AODV for Efficient Power-Aware Routing in MANETs

Boustead, P. 1D-02.4 Rendering Models for Immersive Voice Communications within Distributed Virtual Environment

Boyd, C. 1C-11.3 Characterisation of Combat Identification Technologies

Branch, P. 2B-11.3 Validation of the Random Waypoint Mobility Model Through a Real World Mobility Trace

Branch, P. 2C-11.1 An Architecture for Automated Network Control of QoS over Consumer Broadband Links

Brown, R. 3C-09.4 Using Distributed Circuits to Model Power Capacitor Behavior

Brown, W. 1B-11.4 Neural Network Ensembles Based Approach for Mineral Prospectivity Prediction

Browne, R. 2B-04.1 Bias Characteristics of Bilinear Interpolation Based Registration

Budrikis, Z. 2D-01.3 Internet Transfer Capabilities for Time Critical Communications

Bunder, M. 1C-04.2 Introducing NTS: A Classification System for Knowledge Library Applications

Tencon 2005 Page 143

Name and Initial Session Paper Title

Burnett, I. 1C-04.3 Collaborative Editing using an XML Protocol

Improved Dynamic Multimedia Resource Adaptation-based Peer-to-Peer Sys-tem through Locality-

Burnett, I. 2D-02.3
based Clustering and Service

But, J. 1C-02.2 Evaluating the impact of DNS and HTTP session characteristics on consumer ISP web traffic

C. Liang, Y. 3C-08.5 A Self-contained Direct Methanol Fuel Cell with Surface Tension Fuel Delivery

Cahill, L. 2B-11.11 Optical Switching Using Cascaded Generalised Mach-Zehnder Switches

Cai, J. 1D-08.2 Subjective Image Quality Assessment at the Threshold Level

Cai, L. 1D-02.1 A Late Join Approach for Distributed Dynamic-locking in Real-time Collaborative Editing Systems

Caroline, E. 2D-12.12 Optical Signal Processor for Multibeam array antennas Using Inverse Piezoelectric Effect

Analog Baseband Chain in a 0.18 µm Standard Digital CMOS Technology for IEEE802.15.3a (UWB)
Cha, C. 1C-05.4
Receiver

Chae, C. 2C-11.3 Ethernet PON or WDM PON: A Comparison of Cost and Reliability

Analog Baseband Chain in a 0.18 µm Standard Digital CMOS Technology for IEEE802.15.3a (UWB)
Chae, H. 1C-05.4
Receiver

Chai, Y. 1D-08.5 A New Automatic Gait Recognition method based on the Perceptual Curve

Chaki, N. 3B-05.3 Design Pattern Mining from Source Code for Reverse Engineering

Chakraborty, S. 3B-06.3 Novel Dielectric Rectangular Slab Antenna With Improved Performance

Chamnongthai, K. 3D-10.5 Application of Power Source Element in Power Factor Correction

Chan, S. 2B-06.4 Embedded Web Server Based Power Quality Recorder

Chand, P. 2B-12.8 Co-optimizing Ancillary Raise and Lower Services

Chang, E. 1C-02.4 Context Based Riskiness Assessment

Chang, H. 3B-03.4 Design of DMB Nonuniform SIR-type BPF with Embedded Tunable Pads Using LTCC

Chang, R. 2D-10.3 An Energy Efficient Data Query Architecture for Large Scale Sensor Networks

An Approach to FPGA-based Time-Frequency Spectrogram by Real-Time Sweep Spectral Extraction

Chang, T. 2D-05.1
Algorithm
Use of High-Tc Superconducting Tape RF Coils for Non-invasive Fiber Tracking on Diffusion Tensor
Chang, W. 1C-07.2
MRI application

Chang, W. 3C-09.2 Implementation of Bi-2223 HTS tapes as RF Coils for 3T MRI System

Chang, Y. 2D-10.2 Reliability and Sensitivity Analysis of Embedded Systems with Modular Dynamic Fault Trees

Chang, Y. 1B-05.4 A Robust Design of Reconfigurable Coder/Decoder Against Un-authorized Users

Chang-Chien, L. 3D-03.3 Active Filter Based On-Chip Step-Down DC-DC Switching Voltage Regulator

Chao, Y. 3D-05.4 Skin-based Face Tracking Using Illumination Recognition

Charayaphan, C. 3D-05.3 A Framework for the Design and Implementation of a Dynamic Face Tracking System

Chatterjee, S. 1D-11.2 A Mixed-Split Scheme for 2-D DPCM Based LSF Quantization

Chatterjee, S. 3B-05.3 Design Pattern Mining from Source Code for Reverse Engineering

Chaturvedi, A. 1C-10.1 A Novel Scheme of Load forecasting pertaining to long term planning of a distribution system

Chat-Uthai, C. 3D-10.4 Energy Management Program for Use of Induction Motors Based on Efficiency Prediction

Chat-Uthai, C. 1D-06.4 Minimum Cost Design of Small Low-Loss Transformers

Tencon 2005 Page 144

Name and Initial Session Paper Title

Chaudhuri, A. 1D-10.3 Dynamic User Credential Management in Grid Environment

Chaudhuri, A. 3C-02.3 Privacy Protection of Grid Services in a Collaborative SOA Environment

Che Fung, C. 1C-02.3 Intelligent Customer Relationship Management On The Web

A Nonparametric Discriminant Approach in Resolving Complex Multi-class Query for Content-based

Che Fung, C. 1C-03.4
Image Retrieval

Che Fung, C. 1B-11.4 Neural Network Ensembles Based Approach for Mineral Prospectivity Prediction

Che, L. 1C-01.4 Traffic Measurement for Large Internet Flows Using Multi-stage Landmark LRU Cache

Chen, C. 1D-02.1 A Late Join Approach for Distributed Dynamic-locking in Real-time Collaborative Editing Systems

Chen, C. 3D-09.1 A New Variable Topology Genetic Coding for Evolutionary Hardware

Chen, C. 3D-09.1 A New Variable Topology Genetic Coding for Evolutionary Hardware

Chen, C. 3D-07.2 Design of Limited Buffer Size for Storing Multiple Reference Frames in a Video Codec

Chen, C. 3D-07.2 Design of Limited Buffer Size for Storing Multiple Reference Frames in a Video Codec

Chen, C. 3D-05.1 Face personalization: animated facial modeling approach using radial basis function

Chen, F. 3D-07.2 Design of Limited Buffer Size for Storing Multiple Reference Frames in a Video Codec

Chen, G. 2D-01.1 Impact of IP Traffic on DPSK WDM Systems

Chen, J. 3D-05.4 Skin-based Face Tracking Using Illumination Recognition

Chen, L. 1B-04.2 A Dual-Spiral Reengineering Model for Legacy System

Chen, S. 3D-07.2 Design of Limited Buffer Size for Storing Multiple Reference Frames in a Video Codec

Chen, W. 3B-02.2 Quantitative Availability Prediction for a HA Messaging System

Chen, W. 1C-11.2 An LMI Approach to Exponential Stability Analysis of Neural Networks with Time-Varying Delay

Chen, Y. 2B-06.4 Embedded Web Server Based Power Quality Recorder

Chen, Y. 1D-10.5 An Artificial Neural-Network-Based Approach to Software Reliability Assessment

Cheng, M. 2B-06.4 Embedded Web Server Based Power Quality Recorder

Cheng, Q. 2B-12.5 Blind Channel and Carrier Frequency Offset Estimation for Zero-Prefix OFDM Systems

Cheng, Q. 2B-12.6 Performance of OFDM Carrier Frequency Offset Estimation Based on Multiple Blocks

Chengbing, H. 2C-05.3 M-ary Chirp Spread Spectrum Modulation for Underwater Acoustic Communication

Cheol Hwang, M. 3D-07.3 Fast Intra Prediction Mode Selection Scheme Using Temporal Correlation in H.264

Cheong, J. 1B-08.4 A Semi-automatic System for Measuring Tibial Cartilage Volume

Cherry, A. 2B-11.2 Internet Protocol Television (IPTV)

Chew Thum, P. 2B-12.10 Applications of FACTS Controller for Improving Power Transmission Capability

Chin Kim, G. 2B-06.2 COMTRADE-Based Fault Information System for TNB Substations

Ching Yue, O. 3D-01.2 Measurement-based Performance Model of IP Traffic over 3G Networks

Chinnapichappan, R. 2C-05.4 A Bus Level SystemC Model for Evaluation of Avionics Mission System Data Bus

Chiou, H. 2D-03.4 Extending Cluster File Systems beyond Last Miles

Tencon 2005 Page 145

Name and Initial Session Paper Title

Chiou, L. 3D-03.3 Active Filter Based On-Chip Step-Down DC-DC Switching Voltage Regulator

Chiu, L. 3C-10.5 A High Efficiency Sub-Harmonic Self-Oscillating Mixer for Modern Wireless Communication System

Chiu, L. 3B-01.1 A New Quasi-Lumped Resonant Cell Concept in Modern RF and Microwave Components

A QoS Provisioning Technique in a Residential Gateway using Heterogeneous Network Access

Cho, C. 2C-11.2
Technology

Cho, J. 1C-10.5 Adaptive Setting of Digital Relay for Transmission Line Protection

Chockalingam, A. 3D-01.4 Design and Performance Evaluation of an Inter-System Handover Algorithm in UMTS/GSM Networks

GF(2^191) Elliptic Curve Processor Processor using Montgomery Ladder and High Speed Finite Field
Choi, B. 2C-09.3
Arithmetic Unit
On the Use of Forward Temporal Masking and Cumulative Distribution Mapping for Noisy Speech
Choi, E. 3B-09.2
Recognition

Choi, H. 2C-01.1 An optimum mapping of IPs for On-Chip Network design based on the minimum latency constraint

Choi, H. 1C-05.3 A Coherent Detection-based Symbol Detector algorithm for 2.45GHz LR-WPAN receiver

Choi, H. 2B-05.5 Anti-collision algorithm using Bin slot in RFID System

Choi, I. 1B-11.2 The Detection of Curve-type Defects in the TFT-LCD panels with Machine Vision

Choi, S. 2C-11.9 Performance Evaluation of an Enhanced GDCF

Performance Analysis of Noncoherent UWB Transceiver with PEWB under LOS Residential Channel
Choi, S. 2D-04.2
Environment

Chong Tan, E. 1C-12.5 Output-coding and SVM for multiclass microarray classification

Choo Liaw, H. 2B-12.2 Assembly Modelling Through Constraint-based Manipulations in A Virtual Reality Environment

Choo Liaw, H. 3C-04.3 Robust Control Framework for Piezoelectric Actuation Systems in Micro/Nano Manipulation

Choo Ong, K. 1C-08.2 Bluetooth-enabled ECG Monitoring System

Chow, Y. 3C-02.5 The ARP Virtual Reality System in Addressing Security Threats and Disaster Scenarios

An Approach to FPGA-based Time-Frequency Spectrogram by Real-Time Sweep Spectral Extraction

Chu, S. 2D-05.1
Algorithm

Chuang, C. 3B-07.2 A Novel Statistical Cut-Strategy for DP-based Multiple Biosequence Alignment

Chul Park, Y. 2B-11.6 Performance Improvement of Flooding Algorithm in Highly Mobile Networks

Chung, J. 2B-05.1 Vision Based Motion Tracking System for Interactive Entertainment Applications

Chung, Y. 3C-07.4 A Novel Error Detection and Copyright Protection Scheme for MPEG-2 Video

Chung, Y. 2D-11.2 A study of High Capacity Image Steganographic System

Chusanapiputt, S. 1D-06.1 Minimization of transmission loss with consideration of transformer taps constraint

Chusanapiputt, S. 1D-12.1 Generation Expansion Planning Including Biomass Energy Sources with

Chusanapiputt, S. 1D-12.2 Relative Velocity Updating in Parallel Particle Swarm Optimization Based

Chutchavong, V. 2D-07.1 Equalization of the Distorted Chrominance Signal using IIR Digital Filter with the Bernstein Polynomial

Cicuttini, F. 1B-08.4 A Semi-automatic System for Measuring Tibial Cartilage Volume

Clarke, B. 2B-02.2 Telstra's Fibre To The Premises Network Trial

Coghill, G. 3B-04.3 Online Implementation of Servo Controllers Using Bit-Streams

Conti, P. 1B-09.2 On the Performance of Walsh-LS Codes in CDMA Systems

Tencon 2005 Page 146

Name and Initial Session Paper Title

Cosic, I. 1B-07.2 Development of a conductive photoresist with a mixture of SU-8 and HCl doped polyaniline

Cristoforo, J. 1B-04.2 A Dual-Spiral Reengineering Model for Legacy System

Dakhole, P. 2C-08.1 An Asynchronous BI-Polar Ternary ADC

Dale, M. 2B-03.4 Fast MPLS Network Optimisation using Machine Learning

Daley, G. 1D-01.1 Trust Models and Security Considerations in Multicast Listener Discovery Protocol version 2 (MLDv2)

Das, G. 2D-01.5 Traffic Aggregation in Hybrid Ring-Star Optical Networks

Analysis of Nonlinear Propagation and Wave Mixing of Picosecond Pulses in Semiconductor Optical
Das, N. 1C-07.1
Amplifiers

David, S. 3D-07.5 Improved Algorithms for Implementation of MPEG2 AAC Decoder on FPGA

Davis, S. 1C-04.3 Collaborative Editing using an XML Protocol

Dawson, T. 2B-01.2 On the Periodicity of Time-series Network and Service Metrics

Deng, G. 1D-08.1 Optimizing the FBI Fingerprint Compression Algorithm for BLACKFIN Processor

Derpich, M. 2B-08.5 Probability of Interpolation for a Mute Sample Interpolative A/D Converter with Horizon-Length Two

Dhosi, G. 2B-02.1 Optical Monitoring Issues in Highly Transparent Optical Networks

Diegel, O. 1D-10.4 Intelligent Automated Health Systems for Compliance Monitoring

Dighe, A. 2C-08.1 An Asynchronous BI-Polar Ternary ADC

Dillon, T. 1C-02.4 Context Based Riskiness Assessment

Dimyati, K. 2B-02.3 Scalable AWG-based Multi-Protocol Optical Switch (MPOS) for Future Optical GRID Networking

Do, J. 1C-05.3 A Coherent Detection-based Symbol Detector algorithm for 2.45GHz LR-WPAN receiver

Dods, S. 2B-02.1 Optical Monitoring Issues in Highly Transparent Optical Networks

Dogancay, K. 2C-07.1 LMS Algorithm for Blind Adaptive Nonlinear Compensation

Using Reinforcement Learning for Call Admission Control in Cellular Environments featuring Self-
Dogancay, K. 1C-01.2
Similar Traffic

Doi, J. 2B-05.3 Topologically Robust Surface Reconstruction for 3D Shape Modification

Dong Seok, H. 1B-03.1 A Rotational Motion Estimation Method for Digital Image Stabilization

Dong, C. 3D-08.4 PDSR in Phase Angle Space and SPM Based Security Monitoring of Power Systems

Dong, Y. 2C-07.2 Compressed Space-Time Adaptive Processing (CSTAP)

Dooley, L. 3D-10.1 A Novel Half-Way Shifting Bezier Curve Model

Dreibholz, T. 1C-04.4 The Performance of Reliable Server Pooling Systems in Different Server Capacity Scenarios

Drieberg, M. 2B-12.4 On PAPR Reduction in OFDM Systems: A Technique using Normalized Complex Hadamard Transform

Drieberg, M. 2D-11.11 A New Practical MIMO Channel Estimation Technique for IEEE802.16a: Performance Analysis

Du, X. 3D-09.4 High speed and low area cost FPGA based pattern matching

Du, X. 2D-02.2 Investigation on the Content Popularity Distribution under K-Transformation in Streaming Applications

Du, Y. 2D-12.3 Boiler combustion control using the intelligent controllers

Duc, N. 2C-01.1 An optimum mapping of IPs for On-Chip Network design based on the minimum latency constraint

Tencon 2005 Page 147

Name and Initial Session Paper Title

Duong, H. 1C-01.1 Using Rate Balancing to Improve Throughput in a Wireless Multi-hop Chain Topology

Dutta, J. 1C-07.3 Diode fabricated by layer by layer deposition of semiconductor nanoparticles

Edwards, W. 2D-01.3 Internet Transfer Capabilities for Time Critical Communications

Egan, G. 2C-03.2 Performance of IEEE 802.11b Wireless Links: An Experimental Study

A Survey of FPGAs for Acceleration of High Performance Computing and their Application to
Egan, G. 3D-09.2
Computational Molecular Biology

Ehsani, A. 1C-10.3 An Analytical Method for the Reliability Evaluation of Wind Energy Systems

Einicke, G. 2B-07.3 Discrete-time Smoothing Formulas

A Novel Resource Advertisement and Discovery Model for Ubiquitous Computing Environments using
Ensan, F. 2D-08.4
Mobile Agents
On the Use of Forward Temporal Masking and Cumulative Distribution Mapping for Noisy Speech
Epps, J. 3B-09.2
Recognition

Erke, T. 3C-01.2 Modeling of VoIP Conversational Speech in Noisy Environment

Escoto, M. 1D-12.4 Energy-saving Flyback Converter for EDM Applications

Esselle, K. 3C-06.4 High Gain Compact Circularly Polarized Microstrip Antenna

Eum, S. 2B-01.4 A Failure Analysis of the Tomogravity and EM Methods

Accurate Performance Analysis of Optimum Space Diversity Combining and Equalization in Digital
Fang, H. 1C-12.3
Point-to-point Radio

Fang, Y. 3D-04.3 Real-Time Monitoring and Integrated Automation Control in Steel Rolling Plant

Fardin, E. 1D-04.1 Low cost intergidital BST varactors for tunable microwave applications

Farquharson, M. 1C-06.2 An algorithm for estimating time-varying Doppler at low SNR

Farrell, P. 2B-02.1 Optical Monitoring Issues in Highly Transparent Optical Networks

Faulkner, M. 2B-09.2 Designing for Low ISI in an OFDM Modem

Fechete, A. 2D-12.4 Ozone Sensors based on Layered SAW Devices with: InOx / SiNx / 36º YX LiTaO3 Structure

Feiten, A. 2B-01.1 Power, Rate and QoS Control for Impulse Radio

Feng, D. 1B-03.5 Semi-supervised Segmentation of Textured Images by Using Coupled MRF Model

Feng, H. 2C-03.3 An optimized DCF schema with Downlink and Uplink bandwidth control in 802.11 WLAN

Fengsi, W. 1D-02.3 An Approach for OWL-Based Web Services Components and Composition Framework

Ferra, H. 2B-03.4 Fast MPLS Network Optimisation using Machine Learning

Fitzpatrick, P. 1C-01.3 Hybrid Delay/Blocking Wireless Systems with Greedy Resource Allocation

Fotuhi-Firuzabad, M. 1C-10.3 An Analytical Method for the Reliability Evaluation of Wind Energy Systems

Fraser, D. 2C-04.5 Restoration of Non-Uniformly Warped Images Using a Typical Frame as Prototype

Fu, S. 1B-03.4 Feature-Oriented Coupled Bidirectional Flow for Image Denoising and Edge Sharpening

Fuwa, Y. 3D-02.5 A Secure Printing System for Ubiquitous Network

Ganesan, B. 2B-10.2 Fuzzy Matching Scheme for Stereo Vision based Electronic Travel Aid

Gangadharpalli, S. 2D-05.4 Prominent Region Based Object Recognition using Complex Wavelets

Gao, C. 2D-03.5 The Design and Implementation of a WSRF-Compliant Information Service on Wireless Grids

Tencon 2005 Page 148

Name and Initial Session Paper Title

Gaur, P. 3C-04.1 Sensorless Control Scheme for Permanent Magnet Synchronous Motor

Geng, Y. 1B-03.4 Feature-Oriented Coupled Bidirectional Flow for Image Denoising and Edge Sharpening

Georgievski, M. 2D-02.1 Implementation and Usability of User Interfaces for Quality of Service Management

Ghani, T. 2B-12.9 Development of Power System Analysis Software Using Object Components

Ghazali, O. 2D-09.1 Layered Multicast: A Study of Loss Event Rate Estimation in a Low Level of Statistical Multiplexing

Ghorbani, K. 1D-04.1 Low cost intergidital BST varactors for tunable microwave applications

Ghosh, D. 3C-05.3 Co-articulation Detection in Hand Gestures

Ghosh, D. 1D-03.5 Designing of Human Computer Interactive Platform for Robotic Applications

Ghosh, S. 1B-02.3 A Heuristic for low power QoS routing in ad hoc networks

Giang, T. 2B-11.4 A hybrid admission control scheme for wireless networks

Gilman, A. 2B-04.1 Bias Characteristics of Bilinear Interpolation Based Registration

Gilman, A. 3C-09.5 High-speed Weighing Using Impact on Load Cells

Efficient Utilization of Electrical Energy In Pumping Operations Based on Existing Conditions A Case
Giridhar Kini, P. 3B-04.5
Study

Giridhar Kini, P. 2D-12.11 Importance of Accuracy For Steady State Performance Analysis of 3f Induction Motor

Glickman, M. 1B-10.4 Damping Estimation In Highly Interconnected Power Systems

Godara, L. 2D-07.5 An Efficient Implementation of Generalized Sidelobe Canceler for Broadband Antenna Array Signals

Godfrey, J. 1C-11.3 Characterisation of Combat Identification Technologies

Goodwin, G. 2B-08.5 Probability of Interpolation for a Mute Sample Interpolative A/D Converter with Horizon-Length Two

Gopalakrishnan, S. 2B-10.2 Fuzzy Matching Scheme for Stereo Vision based Electronic Travel Aid

Gopalan, S. 3D-07.1 Clustering based Video Transcoding for Congestion Control in Wireless Environments

Gopalan, S. 2B-11.9 Hybrid P2P Based Self Organising Grids for Efficient Resource Distribution

Gordon, S. 1B-02.5 Ranger, a Novel Intrusion Detection System Architecture for Mobile Ad Hoc Networks

Gordon, S. 1C-01.1 Using Rate Balancing to Improve Throughput in a Wireless Multi-hop Chain Topology

Goshima, T. 3D-02.5 A Secure Printing System for Ubiquitous Network

Gribbon, K. 2D-11.1 Design Patterns for Image Processing Algorithm Development on FPGAs

Guesmi, T. 2B-11.10 Processors Communications in the Context of Virtual Switch Framework Using the CAN-IOP Protocol

Guesmi, T. 2C-12.8 Hard Real Time Communication Using the CANIOP Protocol

Gunawardana, U. 1B-09.2 On the Performance of Walsh-LS Codes in CDMA Systems

Gunawardana, U. 2D-08.2 Monitoring of the Coalesced Ice Ball in Cryosurgery

Guo, D. 1B-06.2 Design and Implementation of Wavelet Filter Using Switched-current Circuits

Guo, G. 2B-09.4 Performance Comparison of Blind Symbol Timing Estimation in Cyclic Prefixed OFDM Systems

Guo, W. 2D-11.10 Data Mining on Patient Data

Guo, W. 1C-02.5 The Design of China Tourism E-Business Platform Based on J2EE Technology

Tencon 2005 Page 149

Name and Initial Session Paper Title

Guo, Y. 1B-02.5 Ranger, a Novel Intrusion Detection System Architecture for Mobile Ad Hoc Networks

Gupta, B. 3B-06.3 Novel Dielectric Rectangular Slab Antenna With Improved Performance

Gupta, S. 2B-04.4 Novel Masks for Multimodality Image Fusion using DTCWT

Gwsice, W. 2D-11.10 Data Mining on Patient Data

Gwsice, W. 1C-02.5 The Design of China Tourism E-Business Platform Based on J2EE Technology

H. Wu, C. 3D-03.3 Active Filter Based On-Chip Step-Down DC-DC Switching Voltage Regulator

Habibi, D. 2B-01.5 Establishing Physical Survivability of Large Networks using Properties of Two-Connected Graphs

Halgamuge, S. 3C-05.5 Using Particle Swarm Optimisation for Elastic Bunch Graph Matching to Recognise Faces

Halgamuge, S. 1D-07.1 Detection of Faulty Semiconductor Wafers using Dynamic Growing Self Organizing Map

Halim Abdul Rashid, A. 2B-12.9 Development of Power System Analysis Software Using Object Components

Hallani, H. 2B-11.1 Wireless Ad-Hoc Networking: Analysis and Validation of Simulation Results

Hämäläinen, T. 2C-01.3 Optimized Pricing and Closed Form Algorithm for WFQ Scheduling

Han, F. 3C-06.1 A Hybrid Method for the Calculation of Radiation Field from Opening in Shielding Enclosure

Han, J. 1C-05.3 A Coherent Detection-based Symbol Detector algorithm for 2.45GHz LR-WPAN receiver

Han, Y. 2C-12.5 Dynamic Spectrum Allocation Using Channel Power in CDMA Networks

Harikumar, R. 3B-07.1 Genetic Algorithm for Classification of Epilepsy Risk Levels from EEG Signals

Harris, R. 2B-01.4 A Failure Analysis of the Tomogravity and EM Methods

Harris, R. 2D-01.4 OptiFlow - A Capacity Management Tool

Harrop, W. 2C-11.7 Transparent IP layer Interception on Enterprise 802.11b/g Networks

Harsha, S. 2B-11.8 Link Layer Mitigation in Rural UHF-MIMO Linking Systems

Haseyama, M. 3C-03.1 A Steady-State Analysis of a CORDIC-Based Adaptive ARMA Lattice Filter

Hashimoto, M. 1C-03.3 A Sampling Switch Design for Liquid Crystal Displays

Hashimoto, N. 2D-10.5 A Self-Distributing Software Environment with Binary Interception for Multi-Projector Displays

Hashizume, H. 1D-09.5 Fast and Accurate Positioning Technique Using Ultrasonic Phase Accordance Method

Hassan, S. 2D-09.1 Layered Multicast: A Study of Loss Event Rate Estimation in a Low Level of Statistical Multiplexing

Hau Lau, C. 2B-11.7 SERA: Service Ratio Based Preemption for Minimizing Throughput Loss in Networks

Hayder Al-Mansoori, M. 1B-07.5 Threshold Features of L-Band Linear Cavity Multiwavelength Brillouin-Erbium Fiber Laser

He, Y. 1B-06.2 Design and Implementation of Wavelet Filter Using Switched-current Circuits

He, Y. 3D-04.5 Robust Stability Analysis of Delayed Neural Networks with Polytopic Type Uncertainties

Hee Soong, B. 2B-11.7 SERA: Service Ratio Based Preemption for Minimizing Throughput Loss in Networks

Heinen, S. 2B-08.2 Calibration Technique for Sigma-Delta Modulation Loops using Radio-Frequency Phase Detection}

Hermansson, G. 3B-01.4 Perceptual Influences on Use of Fixed Lines and Mobile Phones

Hernandez Arieta, A. 1C-08.1 FES as Biofeedback for an EMG Controlled Prosthetic Hand

Tencon 2005 Page 150

Name and Initial Session Paper Title

Hewitt, D. 2B-02.1 Optical Monitoring Issues in Highly Transparent Optical Networks

Higgins, C. 3B-05.2 An On-line Recognition System for Handwritten Pitman Shorthand

Hinton, K. 2B-02.1 Optical Monitoring Issues in Highly Transparent Optical Networks

Hiryanto, L. 2B-03.2 Dynamic Router Tables for Full Expansion/Compression IP Lookup

Performance Analysis of Noncoherent UWB Transceiver with PEWB under LOS Residential Channel
Ho Kim, K. 2D-04.2
Environment

Ho Lee, M. 2D-11.12 The Soft Encrypting Channel Based on Turbo-code En-decoders for Wireless Data Transmission

Ho, Y. 2B-12.3 TFRC Fairness Improving Using the Dynamic Loss Rate Measurement

Hoang, D. 1B-01.5 PIPPON: A Physical Infrastructure-aware P2P Overlay Network

The Development of a Novel Image Quality Metric and a Synthetic Colour Test Image for Objective
Hodgson, B. 1C-03.2
Quality Assessment of Digital Codecs

Holland, A. 1D-04.1 Low cost intergidital BST varactors for tunable microwave applications

Holland, A. 2D-12.4 Ozone Sensors based on Layered SAW Devices with: InOx / SiNx / 36º YX LiTaO3 Structure

Hongjie, J. 1C-10.4 A Real-Time Monitor Framework for Static Voltage Stability of Power System

Use of High-Tc Superconducting Tape RF Coils for Non-invasive Fiber Tracking on Diffusion Tensor
Horng, H. 1C-07.2
MRI application

Horng, H. 3C-09.2 Implementation of Bi-2223 HTS tapes as RF Coils for 3T MRI System

Hosoda, K. 2D-08.5 Macroaction Synthesis for Agent Systems

Hou Chan, C. 3B-01.1 A New Quasi-Lumped Resonant Cell Concept in Modern RF and Microwave Components

Houman, S. 1C-08.3 Human Motion Representation Using Eigenspaces

How Tee, K. 1C-08.2 Bluetooth-enabled ECG Monitoring System

Howarth, J. 3C-06.2 60GHz Radios: Enabling Next-Generation Wireless Applications

Hrushikesha, M. 2D-11.9 How Good is a UML Diagram? A Tool to Check It.

Use of High-Tc Superconducting Tape RF Coils for Non-invasive Fiber Tracking on Diffusion Tensor
Hsieh, C. 1C-07.2
MRI application

Hsieh, C. 3C-09.2 Implementation of Bi-2223 HTS tapes as RF Coils for 3T MRI System

Hsieh, C. 3C-04.2 A Resilient Guaranteed Cost Control Design Via the Robust Two-Stage LQ Reliable Control

Hsieh, G. 1D-11.3 Modeling and design considerations for zvs asymmetrical half-bridge converter

Hsu, A. 1D-07.1 Detection of Faulty Semiconductor Wafers using Dynamic Growing Self Organizing Map

Evaluating the Restorability in WDM Networks Based on Directional Cycle Decomposition Algorithm
Hsu, S. 1B-01.3
(DCDA)

Hu, B. 2C-03.3 An optimized DCF schema with Downlink and Uplink bandwidth control in 802.11 WLAN

Hu, J. 3B-08.1 H-Infinity Design for a Sonar-Platform

Hu, J. 3C-02.2 Developing a Distributed Agent-based Computer Network Security Infrastructure

Hu, Q. 1B-06.2 Design and Implementation of Wavelet Filter Using Switched-current Circuits

Huang, C. 2D-10.2 Reliability and Sensitivity Analysis of Embedded Systems with Modular Dynamic Fault Trees

Huang, C. 2C-05.1 Software Reliability Modeling with Weibull-type Testing-Effort and Multiple Change-Points

Huang, C. 2D-03.4 Extending Cluster File Systems beyond Last Miles

Tencon 2005 Page 151

Name and Initial Session Paper Title

Huang, C. 1D-10.5 An Artificial Neural-Network-Based Approach to Software Reliability Assessment

Huang, C. 2D-03.5 The Design and Implementation of a WSRF-Compliant Information Service on Wireless Grids

Huang, J. 1B-05.4 A Robust Design of Reconfigurable Coder/Decoder Against Un-authorized Users

Huang, J. 3D-09.4 High speed and low area cost FPGA based pattern matching

Huang, J. 2C-05.3 M-ary Chirp Spread Spectrum Modulation for Underwater Acoustic Communication

Huang, X. 2D-01.1 Impact of IP Traffic on DPSK WDM Systems

Hun Kim, D. 2C-11.9 Performance Evaluation of an Enhanced GDCF

Hung, H. 2C-09.1 A new interpolation method for channel estimation of OFDM systems

Hur, J. 3C-07.1 H.264 to MPEG-4 Resolution Reduction Transcoding

Hur, J. 3C-07.2 H.264 to MPEG-4 Transcoding Using Block Type Information

Hussain, O. 1C-02.4 Context Based Riskiness Assessment

Hussain, Z. 2B-09.3 Adaptive Receiver Beamforming for Diversity Coded OFDM Systems: Maximum SNR Design

Hussain, Z. 1B-12.2 Space-Time Encoded Secure Chaos Communications with Transmit Beamforming

Hussain, Z. 1C-05.1 Performance of the Time-Delay Digital Tanlock Loop as PM Demodulator in Additive Gaussian Noise

Hussain, Z. 1D-09.1 Frequency Estimation of Mono- and Multi-component FM Signals Using the T - Distributions

Hussain, Z. 3B-03.3 Efficient Structure for Single-Bit Digital Comb Filters and Resonators

Hussain, Z. 2D-07.3 A Single-Bit Digital DC-Blocker Using Ternary Filtering

Performance of Adaptive Transmit Diversity with Orthogonal Space-Time Block Coding in Microcell and
Hussain, Z. 2B-07.5
Macrocell Channel Environments

Hussain, Z. 1B-09.3 Blind Adaptive Multiuser Detection for DS-CDMA Utilizing Sinusoidally-Distributed DSE-CMA

Transmission of Compressed Multimedia Data over Wireless Channels Using Space-time OFDM with
Hussain, Z. 2B-09.5
Adaptive Beamforming

Hussain, Z. 1C-09.2 Blind Adaptive Multiuser Detection for Chaos CDMA Communication

Hussain, Z. 2D-09.4 A Frequency-Selective Mobile Radio Channel with Hyperbolically Distributed Scatterers

Hussain, Z. 2D-07.4 Spectral Analysis of Delta and Sigma-Delta Modulators Using Linear FM Input Signals

Hussain, Z. 1C-12.4 A Space-Time Block-Coded PCC-OFDM Scheme for Flat Fading Wireless Channels

Performance of a Text-Independent Remote Speaker Recognition Algorithm over Communication

Hussain, Z. 2D-12.5
Channels with Blind Equalisation

Hussain, Z. 2D-12.13 Adaptive LMS Ternary Filtering

Long-haul ASK and DPSK Optical Fibre Transmission Systems: Simulink Modeling and Experimental
Huy Lam, Q. 1C-05.2
Demonstration Test-Beds
Geometrical and Phasor Representation of Multi-level Amplitude-Phase Modulation Formats and
Huynh, T. 1B-05.5
Photonic Transmitter Structures
Long-haul ASK and DPSK Optical Fibre Transmission Systems: Simulink Modeling and Experimental
Huynh, T. 1C-05.2
Demonstration Test-Beds

Hwang, L. 2B-06.4 Embedded Web Server Based Power Quality Recorder

Hyeon-Sun, S. 3B-01.5 Wireless Grid Middleware Architecture for Providing Reliable Services

Hyuk Moon, Y. 3B-01.5 Wireless Grid Middleware Architecture for Providing Reliable Services

A QoS Provisioning Technique in a Residential Gateway using Heterogeneous Network Access

Hyun Jin, L. 2C-11.2
Technology

Tencon 2005 Page 152

Name and Initial Session Paper Title

Ibrahim, Z. 3B-07.4 Direct-Proportional Length-Based DNA Computing Implementation for Elevator Scheduling Problem

Experimental Comparison of Initial Pool Generation for Direct-Proportional Length-Based DNA

Ibrahim, Z. 1B-08.3
Computing

Ilic, V. 2D-08.2 Monitoring of the Coalesced Ice Ball in Cryosurgery

Iqbal, M. 2C-11.8 Optimizing Network Control Traffic in Resource Constrained MANETs

Iqbal, M. 3B-08.2 Adaptive Self Configuration for Distributed Load in Sensor Networks

Ishida, Y. 2D-10.5 A Self-Distributing Software Environment with Binary Interception for Multi-Projector Displays

Ishikawa, S. 2B-04.2 A Mobile Motion Capture System Employing Image Transfer

Ishikawa, S. 1C-08.3 Human Motion Representation Using Eigenspaces

Ishizaki, F. 2C-01.4 Analysis of performance improvement with packet dispersion

Islam, A. 1D-07.1 Detection of Faulty Semiconductor Wafers using Dynamic Growing Self Organizing Map

Islam, M. 1B-02.4 An Intrusion Detection System for Suburban Ad-hoc Networks

Ismail, M. 1B-09.1 Optimization of UTRA SHO Algorithm

Ismail, M. 1B-09.4 Downlink Soft Handover Mechanism Performance in WCDMA System

Ito, T. 3B-06.1 A Reconfigurable RF Circuit Architecture for Dynamic Power Reduction

Ivanovich, M. 1C-01.3 Hybrid Delay/Blocking Wireless Systems with Greedy Resource Allocation

Iwanami, Y. 1C-06.3 A Wavelet Packet Modulation Method for Satellite Communications

Iwanami, Y. 2D-04.3 A higher rate turbo equalization with multilevel block-coded modulation

Iyer, V. 1B-11.4 Neural Network Ensembles Based Approach for Mineral Prospectivity Prediction

Double-Pass Discrete Raman Amplifier with FBG for Pump Power Optimization and Improved
Izyani Md Ali, M. 2C-12.3
Performance
Are the parameters Q0×f0, temperature coefficient of frequency and temperature coefficient of
Jacob, M. 1D-04.3
permittivity fundamental material constants?

Jacob, M. 1D-04.4 Planar Ba(MgxTay)O3 Material for Emerging Microwave Technologies

Jamal, M. 1B-10.3 Industrial Energy Audit using Data Mining Model Web Application

Jamuar, S. 3D-03.4 Applications of Low Voltage FGMOS based CCII in Filtering

Jana, D. 1D-10.3 Dynamic User Credential Management in Grid Environment

Jana, D. 3C-02.3 Privacy Protection of Grid Services in a Collaborative SOA Environment

Janney, P. 2D-05.4 Prominent Region Based Object Recognition using Complex Wavelets

Jantarang, S. 1D-12.1 Generation Expansion Planning Including Biomass Energy Sources with

Jantarang, S. 1D-12.2 Relative Velocity Updating in Parallel Particle Swarm Optimization Based

Jaragh, M. 3D-09.3 Mapping of an N-Way Set Associative Cache to An N-Square Grid for the LRU Algorithm

Jarvis, R. 1D-03.1 Robot Path Planning in High Risk Fire Front Environments

Jarvis, R. 1D-03.3 Accommodating Uncertainty in Covert and Overt Robot Path Planning

Javaid Iqbal, S. 1B-07.5 Threshold Features of L-Band Linear Cavity Multiwavelength Brillouin-Erbium Fiber Laser

Jay Ix Sabido, D. 1B-07.1 Spiral Inductor Coupling on 0.25 um Epitaxial CMOS Process

Tencon 2005 Page 153

Name and Initial Session Paper Title

Jay Ix Sabido, D. 1D-07.4 Monolithic Spiral Inductors for a 0.25 um Digital CMOS Process

Jayaram, D. 3D-01.4 Design and Performance Evaluation of an Inter-System Handover Algorithm in UMTS/GSM Networks

Jea Ko, S. 3D-07.3 Fast Intra Prediction Mode Selection Scheme Using Temporal Correlation in H.264

Jeong Kyoon, L. 2C-11.6 The Implementation of Security Message Protocol for PDA PUSH Service

Jeong Kyoon, L. 2B-12.3 TFRC Fairness Improving Using the Dynamic Loss Rate Measurement

A QoS Provisioning Technique in a Residential Gateway using Heterogeneous Network Access

Jeong, Y. 2C-11.2
Technology

Jeoti, V. 2B-12.4 On PAPR Reduction in OFDM Systems: A Technique using Normalized Complex Hadamard Transform

Jeoti, V. 2D-12.7 Compression and Enhancement of Digital Mammograms with Wavelet Footprints

Jeoti, V. 2D-11.11 A New Practical MIMO Channel Estimation Technique for IEEE802.16a: Performance Analysis

Jha, R. 2C-04.4 Enhancement of Digital Images Using Stochastic Resonance

Ji, P. 1D-02.1 A Late Join Approach for Distributed Dynamic-locking in Real-time Collaborative Editing Systems

Jian, G. 3D-08.3 Three-phase fly-back AC/DC converter with novel resonant snubber circuit

Jiang, B. 1D-02.1 A Late Join Approach for Distributed Dynamic-locking in Real-time Collaborative Editing Systems

Jiang, Y. 2D-06.4 A Visual Attention Model Based on DCT Domain

Jimmy Li, J. 1B-03.3 CFA Demosaicking with Improved Colour Edge Preservation

Jin, H. 2D-02.5 DHT Based Collaborative Multimedia Streaming and Caching Service

Jirasereeamornkul, K. 3D-10.5 Application of Power Source Element in Power Factor Correction

Johnston, C. 2D-11.1 Design Patterns for Image Processing Algorithm Development on FPGAs

Jonkman, M. 2D-05.3 The Application of Wavelet and Feature Vectors to ECG Signals

Joshi, M. 1D-08.3 Fast hybrid fractal-VQ color compression in Wavelet domain

Joutsensalo, J. 2C-01.3 Optimized Pricing and Closed Form Algorithm for WFQ Scheduling

Jung, C. 1C-10.5 Adaptive Setting of Digital Relay for Transmission Line Protection

Jung, M. 2C-11.9 Performance Evaluation of an Enhanced GDCF

Jung-Yeop, J. 3B-01.5 Wireless Grid Middleware Architecture for Providing Reliable Services

Jusak, J. 1B-09.3 Blind Adaptive Multiuser Detection for DS-CDMA Utilizing Sinusoidally-Distributed DSE-CMA

Jusak, J. 1C-09.2 Blind Adaptive Multiuser Detection for Chaos CDMA Communication

Performance of a Text-Independent Remote Speaker Recognition Algorithm over Communication

Jusak, J. 2D-12.5
Channels with Blind Equalisation

K. Moon, Y. 2D-08.3 Design and Implement the Stimuli Capsule at In-vitro Experiment

K. Shyamasundar, R. 2D-11.9 How Good is a UML Diagram? A Tool to Check It.

Kaewarsa, S. 2C-06.2 Power Quality Disturbance Recognition Using Wavelet-Based Neural Networks

A Novel Resource Advertisement and Discovery Model for Ubiquitous Computing Environments using
Kahani, M. 2D-08.4
Mobile Agents

Kalantar-Zadeh, K. 1B-07.2 Development of a conductive photoresist with a mixture of SU-8 and HCl doped polyaniline

Kalantar-Zadeh, K. 2D-12.4 Ozone Sensors based on Layered SAW Devices with: InOx / SiNx / 36º YX LiTaO3 Structure

Tencon 2005 Page 154

Name and Initial Session Paper Title

Kalaratt, K. 2C-04.2 Rapid 3D Simultaneous Algebraic Reconstruction Technique (SART) on Clustering System

Kalia, S. 1B-04.5 Masking approach to secure systems from Operating system Fingerprinting

Kamakura, K. 1C-09.3 Use of Turbo Coding for Embedded Transmission Scheme in Optical CDMA

Kamruzzaman, J. 1B-08.2 Gene Selection and Classification of Human Lymphoma from Microarray Data

Kanazawa, H. 3D-02.5 A Secure Printing System for Ubiquitous Network

Kaneko, A. 1D-09.5 Fast and Accurate Positioning Technique Using Ultrasonic Phase Accordance Method

Kaneko, K. 2C-05.2 Implementation and Evaluation of the Extended Cell Splitting Algorithm

Kang, Q. 2B-10.1 A Fuzzy Adaptive Programming Method of Particle Swarm Optimization

Kanjilal, A. 1B-04.1 Automated Translation of Behavioral Models using OCL and XML

Kannisto, L. 2C-01.3 Optimized Pricing and Closed Form Algorithm for WFQ Scheduling

Development of Efficient Algorithm for Fault Location on Series Compensated Parallel Transmission
Kapuduwage, S. 2C-06.5
Lines

Karim, A. 1D-07.1 Detection of Faulty Semiconductor Wafers using Dynamic Growing Self Organizing Map

Karmakar, G. 3D-10.1 A Novel Half-Way Shifting Bezier Curve Model

Analysis of Nonlinear Propagation and Wave Mixing of Picosecond Pulses in Semiconductor Optical
Karmakar, N. 1C-07.1
Amplifiers

Karmakar, N. 3C-09.3 Quasi-Static Analysis of Defected Ground Structure

Karmakar, N. 3C-03.5 Stepped-Impedance Defected Ground Structure

A Novel Three-Phase Active Power Filter Control Algorithm with Unbalanced and Distorted Supply
Karthikeyan, K. 3D-03.2
Voltages

Kashem, M. 2C-06.4 Maximising Voltage Support in Distribution Systems by Distributed Generation

Kasuga, M. 2C-02.5 Kansei-based investigation of methods for projection of digital content

Kasuga, M. 3D-05.5 Human-Machine Interaction Compensating Imperfection in Facial Expression Recognition

Analysis of Nonlinear Propagation and Wave Mixing of Picosecond Pulses in Semiconductor Optical
Kawaguchi, H. 1C-07.1
Amplifiers

Kawazoe, D. 3B-06.1 A Reconfigurable RF Circuit Architecture for Dynamic Power Reduction

Kazimierczuk, M. 3D-10.5 Application of Power Source Element in Power Factor Correction

Kazuya, M. 3B-06.1 A Reconfigurable RF Circuit Architecture for Dynamic Power Reduction

Kedsoi, T. 3D-10.4 Energy Management Program for Use of Induction Motors Based on Efficiency Prediction

Keller, U. 1C-02.2 Evaluating the impact of DNS and HTTP session characteristics on consumer ISP web traffic

Kennedy, D. 2D-03.1 Evaluation of FAST TCP in Low-Speed DOCSIS-based Access Network

Keogh, D. 3D-06.5 The Capacity of Two Deterministic MIMO Radio Channels

Khadeer Hussain, F. 1C-02.4 Context Based Riskiness Assessment

Grid-connected Photovoltaic System with Maximum Power Point Tracking using Self-Organizing Fuzzy
Khaehintung, N. 2C-10.3
Logic Controller
A Solar-powered Battery Charger with Neural Network Maximum Power Point Tracking Implemented
Khaehintung, N. 1B-07.3
on a Low-Cost PIC-microcontroller

Khalid, M. 3B-07.4 Direct-Proportional Length-Based DNA Computing Implementation for Elevator Scheduling Problem

Experimental Comparison of Initial Pool Generation for Direct-Proportional Length-Based DNA

Khalid, M. 1B-08.3
Computing

Tencon 2005 Page 155

Name and Initial Session Paper Title

Khalsa, H. 3C-08.4 Choice of Reference Conductor in Three Phase Systems A Paradigm Shift

Khalsa, H. 2B-06.5 Three Phase Generalised Power Theory for the Measurement of Powers in an Electric Power System

Khambete, M. 1D-08.3 Fast hybrid fractal-VQ color compression in Wavelet domain

Generation of Excitation Signal in Voice excited Linear Predictive Coding using Discrete Cosine
Khare, G. 1D-05.1
Transform

Khazani Abdullah, M. 1B-07.5 Threshold Features of L-Band Linear Cavity Multiwavelength Brillouin-Erbium Fiber Laser

Kho, D. 1C-08.2 Bluetooth-enabled ECG Monitoring System

Performance Analysis of Noncoherent UWB Transceiver with PEWB under LOS Residential Channel
Ki, M. 2D-04.2
Environment

Kia Eng, G. 3B-09.3 Applying Hybrid Neural Network For Malay Syllables Speech Recognition

Kikkert, C. 3C-03.3 Designing Low Cost Wideband Microstrip Bandpass Filters

Kikkert, C. 1D-04.5 Dual Coupled Resonator Local Oscillator

Kim, H. 3B-08.3 Semi-Continueous Interpolation Algorithm for Aspherical Surface Grinding

Kim, J. 1C-10.5 Adaptive Setting of Digital Relay for Transmission Line Protection

Kim, J. 3C-07.5 A Scheduling Method for Real-time Decoding of MPEG-4 Video Streams

Kim, J. 2B-05.5 Anti-collision algorithm using Bin slot in RFID System

A QoS Provisioning Technique in a Residential Gateway using Heterogeneous Network Access

Kim, J. 2C-11.2
Technology

Kim, K. 2B-07.2 A Linear Fusion of the Local Kalman Estimates

Kim, K. 2C-12.2 PWM Inverter IP for High-Fidelity Audio Applications

Kim, R. 2C-12.2 PWM Inverter IP for High-Fidelity Audio Applications

Kim, T. 2B-01.3 Design of an Enhanced Fairness Algorithm for the IEEE 802.17 Resilient Packet Ring

Kim, W. 2C-07.3 Adaptive Surface Fitting for Inspection of FPD Devices Using Multilevel B-spline Approximation

Kim, W. 1B-11.2 The Detection of Curve-type Defects in the TFT-LCD panels with Machine Vision

Analysis and Modeling of the Priority Serving Scheme (PSS) and the Efficient Utilisation of
King, H. 2D-09.2
Communication Channels

Kininmonth, S. 1D-04.2 High Speed Over Ocean Radio Link to Great Barrier Reef

Kiong Nguang, S. 3B-04.3 Online Implementation of Servo Controllers Using Bit-Streams

Kiranon, W. 3B-03.1 Electronically current mode all-pass filter with only one grounded capacitor

A Solar-powered Battery Charger with Neural Network Maximum Power Point Tracking Implemented
Kiranon, W. 1B-07.3
on a Low-Cost PIC-microcontroller

Kishore, B. 2D-11.9 How Good is a UML Diagram? A Tool to Check It.

Kist, A. 2D-01.4 OptiFlow - A Capacity Management Tool

Kitajima, H. 3C-03.1 A Steady-State Analysis of a CORDIC-Based Adaptive ARMA Lattice Filter

Ko, H. 3B-03.4 Design of DMB Nonuniform SIR-type BPF with Embedded Tunable Pads Using LTCC

Analog Baseband Chain in a 0.18 µm Standard Digital CMOS Technology for IEEE802.15.3a (UWB)
Koh, J. 1C-05.4
Receiver

Koilpillai, D. 2D-04.4 Equivalence of the Linear Symbol-level MMSE and the Generalized RAKE

Impact of Propagation Parameters on Neighbouring Cells Interference in Power Controlled CDMA

Kolahi, S. 1C-09.4
Systems

Tencon 2005 Page 156

Name and Initial Session Paper Title
Semantic Indexing of News Video Sequences: A Multimodal Hierarchical approach based on Hidden
Kolekar, M. 3D-07.4
Markov Model

Komatsu, K. 3D-02.5 A Secure Printing System for Ubiquitous Network

Konda, L. 2D-11.9 How Good is a UML Diagram? A Tool to Check It.

Automatic Offset Calibration of Quadrature Encoder Pulse Sensor For Vector Controlled Drive of
Konghirun, M. 2B-12.7
Permanent Magnet Synchronous Motors

Kooi Tan, J. 2B-04.2 A Mobile Motion Capture System Employing Image Transfer

Kooi Tan, J. 1C-08.3 Human Motion Representation Using Eigenspaces

Kopp, C. 1B-02.4 An Intrusion Detection System for Suburban Ad-hoc Networks

Koseeyaporn, J. 1C-07.5 A Novel Monostable Multivibrator Circuit

Koseeyaporn, J. 1D-09.4 Human Identification System Based ECG Signal

Kouzani, A. 3C-05.1 Attractor-Based Person Identification

Krause, D. 1C-11.3 Characterisation of Combat Identification Technologies

Krittayanun, B. 2B-07.4 Improved Performance for Stereophonic Acoustic Echo Cancellation using Subband Structure

Kruth, A. 2B-08.2 Calibration Technique for Sigma-Delta Modulation Loops using Radio-Frequency Phase Detection}

Kuen Kwak, H. 1B-03.1 A Rotational Motion Estimation Method for Digital Image Stabilization

Generation of Excitation Signal in Voice excited Linear Predictive Coding using Discrete Cosine
Kulkarni, M. 1D-05.1
Transform

Kumar Biswas, P. 2C-04.4 Enhancement of Digital Images Using Stochastic Resonance

Kumar Bora, P. 3C-05.3 Co-articulation Detection in Hand Gestures

Kumar Bora, P. 1D-03.5 Designing of Human Computer Interactive Platform for Robotic Applications

Kumar Chakka, V. 2C-03.4 Enhancing Data Efficiency in IEEE- 802.11a Using Frequency and Time domain Equalization

Kumar Cherupalli, P. 1C-06.1 A Wavelet Based Feature Extraction For Voice Lock Systems

Kumar Pusukuri, K. 1C-04.1 Applying Machine Learning Techniques to improve Linux Process Scheduling

Grid-connected Photovoltaic System with Maximum Power Point Tracking using Self-Organizing Fuzzy
Kunakorn, A. 2C-10.3
Logic Controller
A Solar-powered Battery Charger with Neural Network Maximum Power Point Tracking Implemented
Kunakorn, A. 1B-07.3
on a Low-Cost PIC-microcontroller

Kunimune, H. 3D-02.5 A Secure Printing System for Ubiquitous Network

Recovery from Segmentation Failures Using Photometric Invariance in an Interactive Object

Kuno, Y. 1D-03.2
Recognition System

Kuo, C. 1D-11.3 Modeling and design considerations for zvs asymmetrical half-bridge converter

Use of High-Tc Superconducting Tape RF Coils for Non-invasive Fiber Tracking on Diffusion Tensor
Kuo, L. 1C-07.2
MRI application

Kuo, L. 3C-09.2 Implementation of Bi-2223 HTS tapes as RF Coils for 3T MRI System

Recovery from Segmentation Failures Using Photometric Invariance in an Interactive Object

Kurnia, R. 1D-03.2
Recognition System

Kurup, G. 1D-01.1 Trust Models and Security Considerations in Multicast Listener Discovery Protocol version 2 (MLDv2)

Kwok, D. 2D-10.4 A database (DB) parallel processing for computational physics

Kyo Park, S. 3B-03.4 Design of DMB Nonuniform SIR-type BPF with Embedded Tunable Pads Using LTCC

Lambert, A. 2C-04.5 Restoration of Non-Uniformly Warped Images Using a Typical Frame as Prototype

Tencon 2005 Page 157

Name and Initial Session Paper Title

Lan, H. 3B-04.1 Robust Nonlinear Coordinated Control of TCSC and Generator Excitation

Le, A. 2C-06.4 Maximising Voltage Support in Distribution Systems by Distributed Generation

Le, H. 1B-01.5 PIPPON: A Physical Infrastructure-aware P2P Overlay Network

Performance of a Text-Independent Remote Speaker Recognition Algorithm over Communication

Lech, M. 2D-12.5
Channels with Blind Equalisation

Ledwich, G. 3C-08.2 Rapid Detection of Changes to Individual Modes in Multimodal Power Systems

Ledwich, G. 1C-06.2 An algorithm for estimating time-varying Doppler at low SNR

Ledwich, G. 1B-10.4 Damping Estimation In Highly Interconnected Power Systems

Ledwich, G. 2C-06.4 Maximising Voltage Support in Distribution Systems by Distributed Generation

Lee, B. 2B-02.3 Scalable AWG-based Multi-Protocol Optical Switch (MPOS) for Future Optical GRID Networking

Evaluating the Restorability in WDM Networks Based on Directional Cycle Decomposition Algorithm
Lee, C. 1B-01.3
(DCDA)

Lee, C. 1B-05.4 A Robust Design of Reconfigurable Coder/Decoder Against Un-authorized Users

Lee, G. 2C-07.3 Adaptive Surface Fitting for Inspection of FPD Devices Using Multilevel B-spline Approximation

Analog Baseband Chain in a 0.18 µm Standard Digital CMOS Technology for IEEE802.15.3a (UWB)
Lee, H. 1C-05.4
Receiver

Lee, H. 3B-09.4 Sub-Nyquist Nonuniform Sampling and Perfect Reconstruction of Speech Signals

Lee, J. 2C-11.10 A New Cellular Search Algorithm For Motion Estimation

Analog Baseband Chain in a 0.18 µm Standard Digital CMOS Technology for IEEE802.15.3a (UWB)
Lee, J. 1C-05.4
Receiver

Lee, J. 2C-07.3 Adaptive Surface Fitting for Inspection of FPD Devices Using Multilevel B-spline Approximation

Lee, J. 2D-03.4 Extending Cluster File Systems beyond Last Miles

Lee, J. 2D-08.3 Design and Implement the Stimuli Capsule at In-vitro Experiment

Lee, M. 2C-12.2 PWM Inverter IP for High-Fidelity Audio Applications

Lee, P. 2C-10.1 Fuzzy Logic Based Error Concealment Algorithm for H.264 Video Transmission

Lee, S. 2B-02.5 Cost Model based Configuration Management Policy in OBS Networks

Lee, S. 2C-09.1 A new interpolation method for channel estimation of OFDM systems

Lee, T. 2C-11.9 Performance Evaluation of an Enhanced GDCF

Lee, T. 1C-05.5 Stage-Hopping Memory Management Technique for Viterbi Decoders

Lee, U. 1C-05.3 A Coherent Detection-based Symbol Detector algorithm for 2.45GHz LR-WPAN receiver

Lee, Y. 3C-07.3 MPEG-4 to H.264 Transcoding

Lee, Y. 1D-10.1 Reconfigurable Computing: Peripheral Power and Area Optimisation Techniques

Lee, Y. 2C-11.9 Performance Evaluation of an Enhanced GDCF

Leeddham, G. 3B-05.2 An On-line Recognition System for Handwritten Pitman Shorthand

Leeddham, G. 3C-05.4 Document examiner feature extraction: thinned vs. skeletonised handwriting images

Lei, C. 2D-03.4 Extending Cluster File Systems beyond Last Miles

Leiwo, J. 3D-02.1 Strategies for cost efficient security evaluations

Tencon 2005 Page 158

Name and Initial Session Paper Title

Leou, R. 1B-10.2 A Price-Based Unit Commitment Model Considering Uncertainties

Leung Lau, K. 3C-06.5 A Wideband Circularly Polarized Patch Antenna with Conical Radiation Pattern

Leung Lau, K. 3D-06.3 A Wideband Circularly Polarized L-Probe Coupled Patch Antenna for Dual-Band Operation

Lew, D. 2B-11.5 Application of Assisted TDOA Technology in Vehicle Positioning and Navigation System

Leyonhjelm, S. 2B-09.2 Designing for Low ISI in an OFDM Modem

Li, H. 1B-06.2 Design and Implementation of Wavelet Filter Using Switched-current Circuits

Li, H. 1C-10.4 A Real-Time Monitor Framework for Static Voltage Stability of Power System

Li, J. 1C-10.2 The enactment of optimal target value in power plant operation based on language association rule

Li, J. 2B-10.4 The Application of Fuzzy Data Mining in Coal-Fired Boiler Combustion System

Li, L. 2D-11.4 Two-phases Parallel Neural Network Algorithm Based on RPROP

Li, M. 2D-03.5 The Design and Implementation of a WSRF-Compliant Information Service on Wireless Grids

Li, X. 2D-02.4 Design and Implementation of Stream Partitioning Strategies in a Distributed Multimedia System

Liao, G. 2C-08.4 A Novel Maximum Likelihood Frequency Synchronizer for OFDM Systems

Using Reinforcement Learning for Call Admission Control in Cellular Environments featuring Self-
Lilith, N. 1C-01.2
Similar Traffic

Lim, B. 3B-03.4 Design of DMB Nonuniform SIR-type BPF with Embedded Tunable Pads Using LTCC

Lin, B. 3C-10.1 Three-Phase Capacitor-Clamped Active Power Filter for Harmonic and Reactive Power Compensation

Lin, B. 3C-10.3 Series active power filter for current harmonic and load voltage compensation

Analysis, Design and Implementation of an Active Clamp Forward Converter with Synchronous
Lin, B. 3C-09.1
Rectifier

Lin, C. 3C-04.2 A Resilient Guaranteed Cost Control Design Via the Robust Two-Stage LQ Reliable Control

Lin, C. 2C-05.1 Software Reliability Modeling with Weibull-type Testing-Effort and Multiple Change-Points

Lin, C. 3D-05.4 Skin-based Face Tracking Using Illumination Recognition

Lin, G. 1C-05.5 Stage-Hopping Memory Management Technique for Viterbi Decoders

Accurate Performance Analysis of Optimum Space Diversity Combining and Equalization in Digital
Lin, H. 1C-12.3
Point-to-point Radio
Use of High-Tc Superconducting Tape RF Coils for Non-invasive Fiber Tracking on Diffusion Tensor
Lin, I. 1C-07.2
MRI application

Lin, I. 3C-09.2 Implementation of Bi-2223 HTS tapes as RF Coils for 3T MRI System

Lin, K. 2B-09.3 Adaptive Receiver Beamforming for Diversity Coded OFDM Systems: Maximum SNR Design

Lin, K. 1B-12.2 Space-Time Encoded Secure Chaos Communications with Transmit Beamforming

Performance of Adaptive Transmit Diversity with Orthogonal Space-Time Block Coding in Microcell and
Lin, K. 2B-07.5
Macrocell Channel Environments
Transmission of Compressed Multimedia Data over Wireless Channels Using Space-time OFDM with
Lin, K. 2B-09.5
Adaptive Beamforming
An Approach to FPGA-based Time-Frequency Spectrogram by Real-Time Sweep Spectral Extraction
Lin, M. 2D-05.1
Algorithm

Lin, M. 2C-10.1 Fuzzy Logic Based Error Concealment Algorithm for H.264 Video Transmission

Accurate Performance Analysis of Optimum Space Diversity Combining and Equalization in Digital
Lin, S. 1C-12.3
Point-to-point Radio

Lin, Y. 3D-05.4 Skin-based Face Tracking Using Illumination Recognition

Tencon 2005 Page 159

Name and Initial Session Paper Title

Ling, R. 3C-03.5 Stepped-Impedance Defected Ground Structure

Liu, C. 2D-12.2 A Kind of Adaptive Genetic Algorithm and Application in Nonlinear Model Identification

Liu, C. 2B-06.3 Structure Design For Real-Time Thermal Performance Analysis And Monitoring System Of Power Plant

Liu, F. 2D-04.1 A new Modulation Scheme for GSM EDGE

Liu, H. 2C-08.4 A Novel Maximum Likelihood Frequency Synchronizer for OFDM Systems

Liu, M. 1B-06.2 Design and Implementation of Wavelet Filter Using Switched-current Circuits

Liu, Q. 3B-02.4 On Congestion Control in Multicast Networks

Liu, Q. 1D-06.3 A Novel Inductance Calculation Method in Power Transformer Model Based on Magnetic Circuit

Liu, Q. 1B-10.5 A Novel Method on Substation Reactive Power Control with the Fuzzy Logic Method

Liu, W. 1B-08.1 Mapping of Genetic Algorithms for Protein Foilding onto Computational Grids

Liu, W. 3D-09.4 High speed and low area cost FPGA based pattern matching

Liu, W. 2D-02.2 Investigation on the Content Popularity Distribution under K-Transformation in Streaming Applications

Liu, Y. 2D-04.5 Bit-interleaving in fast FH/SSMA systems with convolutional codes and turbo codes

Liu, Z. 3D-01.5 On Hermitian Mapping for Direction-of-Arrival Estimation of Coherent Cyclostationary Signals

Liyana-Pathirana, R. 1B-09.2 On the Performance of Walsh-LS Codes in CDMA Systems

Liyana-Pathirana, R. 2D-08.2 Monitoring of the Coalesced Ice Ball in Cryosurgery

Lizier, J. 2B-01.2 On the Periodicity of Time-series Network and Service Metrics

Lo, H. 2D-10.2 Reliability and Sensitivity Analysis of Embedded Systems with Modular Dynamic Fault Trees

An Algorithm to Allocate the Testing-Effort Expenditures Based on Sensitive Analysis Method for
Lo, J. 1B-04.3
Software Module Systems

Lo, K. 2B-01.5 Establishing Physical Survivability of Large Networks using Properties of Two-Connected Graphs

Detection Schemes for Distributed Space-Time Block Coding in Time-Varying Wireless Cooperative
Lok, T. 1B-12.3
Systems

Lok, T. 1B-12.4 Performance Analysis of Multiple-Relay Decode-and-Forward Cooperation System

Look Law, C. 1B-02.1 CLA-QOS: A Cross-Layer QoS Provisioning Approach for Mobile Ad-hoc Networks

Loong Chong, F. 3B-09.1 A Methodology for Improving PESQ Accuracy for Chinese Speech

Lu, J. 2C-02.3 Quantified Cost-Balanced Overlay Multicast Routing for Multimedia Network

Sustainable Agriculture Mobile Commerce (SamCom) Extending the rural grocery supply chain through
Lu, N. 3C-01.3
M-Commerce

Lu, Y. 2C-05.2 Implementation and Evaluation of the Extended Cell Splitting Algorithm

Luehdorff, H. 1B-09.3 Blind Adaptive Multiuser Detection for DS-CDMA Utilizing Sinusoidally-Distributed DSE-CMA

Luk, K. 3C-06.3 A Patch Loaded Dielectric Resonator Antenna

Lum Tan, W. 3D-01.2 Measurement-based Performance Model of IP Traffic over 3G Networks

Partial Discharges Multi-Targets Localization in Power Transformers Based On Array Signal

Luo, R. 1B-12.1
Processing

Luo, S. 3D-04.2 Greedy exploration policy of Q-learning based on state balance

Lyou, J. 1B-03.1 A Rotational Motion Estimation Method for Digital Image Stabilization

Tencon 2005 Page 160

Name and Initial Session Paper Title

Lyul Lee, Y. 3C-07.1 H.264 to MPEG-4 Resolution Reduction Transcoding

Lyul Lee, Y. 3C-07.2 H.264 to MPEG-4 Transcoding Using Block Type Information

Lyul Lee, Y. 3C-07.3 MPEG-4 to H.264 Transcoding

M. Eshanta, O. 1B-09.1 Optimization of UTRA SHO Algorithm

A Novel Wide-Swing Wide-Bandwidth Scalable Low-Voltage Analog CMOS Multiplier for

M. Hasan, S. 2B-08.4
Communication Signal Processing

Ma, J. 1D-06.2 Research on Novel Power Transformer Protection Based on Self-Correction Function

Ma, W. 2B-12.2 Assembly Modelling Through Constraint-based Manipulations in A Virtual Reality Environment

Ma, Y. 3B-05.2 An On-line Recognition System for Handwritten Pitman Shorthand

Performance of Adaptive Transmit Diversity with Orthogonal Space-Time Block Coding in Microcell and
Mahmoud, S. 2B-07.5
Macrocell Channel Environments

Mahmud, R. 1D-07.2 Frequency and Flexibility Based Cell Phone Keypad Layout

Makinouchi, A. 2C-05.2 Implementation and Evaluation of the Extended Cell Splitting Algorithm

Malik, M. 2D-07.2 State-Space Least Mean Square with Adaptive Memory

Mamaani Barnaghi, P. 2C-02.1 Ontology-Based Multimedia Presentation Generation

Mamaani Barnaghi, P. 2C-11.5 Bluetooth Network Security: A New Approach to Secure Scatternet Formation

Mamun, Q. 3D-04.1 Timestamp based optimistic concurency control

Mangotra, L. 3D-03.4 Applications of Low Voltage FGMOS based CCII in Filtering

Mansoor, S. 1D-03.4 Bluetooth Based Proximity Sensing for Reactive Mobile Robots

Maricar, N. 1B-10.3 Industrial Energy Audit using Data Mining Model Web Application

Marik, M. 3D-09.5 Logic Synthesis and Technology Mapping of MUX-based

Martin, R. 1D-08.2 Subjective Image Quality Assessment at the Threshold Level

Maruyama, Y. 3D-02.5 A Secure Printing System for Ubiquitous Network

Marzouqi, M. 1D-03.1 Robot Path Planning in High Risk Fire Front Environments

Marzouqi, M. 1D-03.3 Accommodating Uncertainty in Covert and Overt Robot Path Planning

Maskell, D. 1D-10.1 Reconfigurable Computing: Peripheral Power and Area Optimisation Techniques

Maskell, D. 1B-05.3 Improving Group Delay Measurement Accuracy using the FM Envelope Delay Technique

Maskell, D. 3D-02.1 Strategies for cost efficient security evaluations

Mathar, R. 2B-01.1 Power, Rate and QoS Control for Impulse Radio

Matsumoto, M. 2C-01.2 Block Based Window Retransmission ARQ Scheme for 100Mbit/s Infrared Links

Matsuoka, K. 2D-04.3 A higher rate turbo equalization with multilevel block-coded modulation

Matsuura, T. 2D-06.1 Determine Calibration Parameters with Satisfied Constraints for Coplanar Camera Calibration

Matsuyama, A. 2D-05.3 The Application of Wavelet and Feature Vectors to ECG Signals

Mazierska, J. 1D-04.4 Planar Ba(MgxTay)O3 Material for Emerging Microwave Technologies

Mcloughlin, I. 2B-08.1 A scalable parallel computational core for embedded processing

Tencon 2005 Page 161

Name and Initial Session Paper Title

Mcloughlin, I. 3B-09.1 A Methodology for Improving PESQ Accuracy for Chinese Speech

Mcloughlin, I. 2B-11.8 Link Layer Mitigation in Rural UHF-MIMO Linking Systems

Mcnickle, D. 2D-01.2 Comparing different approaches to the use of DiffServ on the Internet

Meher, P. 2C-08.2 A New Convolutional Formulation of the DFT and Efficient Systolic Implementation

Mei, L. 2D-12.9 Visual Assessment for The Quantization Error in Wavelet Based Monochrome Videos

Mekhilef, S. 1D-11.4 Digital Control of High DC Voltage Converter Based on Cockcroft Walton Voltage Multiplier Circuit

Mekhilef, S. 3C-10.2 A New Approach for Harmonic Compensation Using Single- phase Hybrid Active Power filter

Melkonian, M. 3B-06.5 ZWOS A Lightweight Operating System For Lightweight Embedded Wireless Devices

Identification of Nonlinear Differential Equation Models from Generalized Frequency Response

Mendes, E. 3B-03.2
Functions

Mercankosk, G. 2D-01.3 Internet Transfer Capabilities for Time Critical Communications

Miao, Z. 2D-10.1 Discussion on Pervasive Computing Paradigm

Michael Odulio, C. 1D-12.4 Energy-saving Flyback Converter for EDM Applications

Min Gu, G. 2B-03.1 Improved Vector Route Algorithm Bases on Baster Topological Space Model

Min Yew, K. 2B-12.4 On PAPR Reduction in OFDM Systems: A Technique using Normalized Complex Hadamard Transform

Min Yew, K. 2D-11.11 A New Practical MIMO Channel Estimation Technique for IEEE802.16a: Performance Analysis

Min, S. 1C-05.3 A Coherent Detection-based Symbol Detector algorithm for 2.45GHz LR-WPAN receiver

Minh Trung, T. 3B-01.5 Wireless Grid Middleware Architecture for Providing Reliable Services

Minhas, R. 2B-07.2 A Linear Fusion of the Local Kalman Estimates

Mir-Nasiri, N. 2B-12.1 Camera-based 3D Tracking System

Mishra, M. 3D-03.1 A Fast Transient Response Single Phase Active Power Filter

A Novel Three-Phase Active Power Filter Control Algorithm with Unbalanced and Distorted Supply
Mishra, M. 3D-03.2
Voltages

Mishra, S. 2C-06.1 Intelligent Classification of Power Quality Disturbances in a Distribution Network

Miyahara, T. 2D-08.5 Macroaction Synthesis for Agent Systems

Miyahara, T. 3B-05.4 A Study on Face Similarity Using Knowledge Processing Methods

Mohamed Nor, K. 3C-08.3 Thermal Analysis of a Tubular Permanent Magnet Linear Generator Using Multiphysics Solver

Mohamed Nor, K. 2B-12.9 Development of Power System Analysis Software Using Object Components

Mohamed Nor, K. 2B-05.2 Transient Stability Program Using Component-Based Software Engineering

Mohamed, N. 3B-01.2 Modification of An Open Area Okumura-Hata Propagation Model Suitable For Oman

Mohammad Ranjbar, A. 1C-10.3 An Analytical Method for the Reliability Evaluation of Wind Energy Systems

Mohammadi Tabari, N. 2B-12.11 A New Approach for Generation Maintenance Scheduling in a Deregulated Power System

Mohammadi, M. 3C-08.1 Model of corona for transient study

Mohd Aripin, N. 1B-09.4 Downlink Soft Handover Mechanism Performance in WCDMA System

Mohd. Ali, B. 3D-06.4 A Simplified Sphere Decoding Algorithm for MIMO Transmission System

Tencon 2005 Page 162

Name and Initial Session Paper Title

Mohd. Ali, B. 2B-09.1 Full Rate Concatenated Space-Frequency and Space-Time OFDM over Naftali Fading Channels

Mohd. Daud, S. 1B-06.3 Comparison of few wavelets to filter ocular artifacts in EEG using Lifting Wavelet Transform

Mokhlis, H. 2B-12.9 Development of Power System Analysis Software Using Object Components

Analysis of Real-Time Multiple Source Video Streaming via Wireless 5.8 GHz for Intelligent Traffic
Monn Baskaran, V. 1B-01.4
Management System

Moon, S. 1B-05.2 Exact Closed-Form Performance Analysis of MFSK Demodulator Based on FFT

Moors, T. 1C-02.1 Towards Robust Internet Applications: Self-Organizing Overlays

Mori, T. 2C-02.5 Kansei-based investigation of methods for projection of digital content

Mossop, D. 3D-02.2 Security Models in the Password-Capability System

Mueller-Wittig, W. 3B-07.3 Using Graphics Hardware to Accelerate Biological Sequence Analysis

Muhammad Sheikh, N. 3C-01.1 Performance Analysis of Adaptive Source Rate Control Algorithm (ASRC) for VoIP

Muhammad, M. 3C-10.2 A New Approach for Harmonic Compensation Using Single- phase Hybrid Active Power filter

Mukundan, R. 3B-05.1 Radial Tchebichef Invariants for Pattern Recognition

Murphy, J. 2B-01.4 A Failure Analysis of the Tomogravity and EM Methods

Murshed, M. 1C-03.1 A motion-based approach for temporal texture synthesis

Murthy, M. 1C-10.1 A Novel Scheme of Load forecasting pertaining to long term planning of a distribution system

Murugesan, P. 1B-11.5 Establishment of Community Information Network in a Developing Nation

Myo Hywe, S. 3B-05.2 An On-line Recognition System for Handwritten Pitman Shorthand

N. Chatterji, B. 2C-04.4 Enhancement of Digital Images Using Stochastic Resonance

N. Jafri, M. 2D-12.10 Target Tracking in an Image Sequence Using Wavelet Features and a Neural Network

Efficient Utilization of Electrical Energy In Pumping Operations Based on Existing Conditions A Case
N. Sreedhar, P. 3B-04.5
Study

N. Sreedhar, P. 2D-12.11 Importance of Accuracy For Steady State Performance Analysis of 3f Induction Motor

Nagargadde, A. 2B-11.9 Hybrid P2P Based Self Organising Grids for Efficient Resource Distribution

A Novel Resource Advertisement and Discovery Model for Ubiquitous Computing Environments using
Naghibzadeh, M. 2D-08.4
Mobile Agents

Naik, S. 3D-07.5 Improved Algorithms for Implementation of MPEG2 AAC Decoder on FPGA

Recovery from Segmentation Failures Using Photometric Invariance in an Interactive Object

Nakamura, A. 1D-03.2
Recognition System

Nakazato, H. 3D-04.1 Timestamp based optimistic concurency control

Nantivatana, P. 2D-12.8 Multiresolution Image Alignment based on Discrete Wavelet Transform

Naraki, T. 2D-08.5 Macroaction Synthesis for Agent Systems

Narasimhan, L. 1B-04.4 A BDL-Based E-Commerce Transaction Architecture

Narasimhan, L. 2D-11.6 Algebraic Semantics for Complete Interaction Sequence

Narkbuakaew, W. 2D-11.3 3D Modeling from Using Spiral Cone-Beam Trajectory

Nasimuddin, N. 3C-06.4 High Gain Compact Circularly Polarized Microstrip Antenna

Nasir, Q. 2C-08.3 Extended Locking Range DTL using Chaos Technique

Tencon 2005 Page 163

Name and Initial Session Paper Title

Neelanarayanan, N. 1B-11.5 Establishment of Community Information Network in a Developing Nation

Negi, A. 1C-04.1 Applying Machine Learning Techniques to improve Linux Process Scheduling

Negnevitsky, M. 2C-06.4 Maximising Voltage Support in Distribution Systems by Distributed Generation

Performance of a Text-Independent Remote Speaker Recognition Algorithm over Communication

Neville, K. 2D-12.5
Channels with Blind Equalisation

Neyer, A. 2B-08.2 Calibration Technique for Sigma-Delta Modulation Loops using Radio-Frequency Phase Detection}

Nezami, M. 2C-08.5 Carrier Recovery Algorithm for Bandwidth-Efficient Coded Modulations at Low Signal-to-Noise Ratios

A Solar-powered Battery Charger with Neural Network Maximum Power Point Tracking Implemented
Ngamkhak, M. 1B-07.3
on a Low-Cost PIC-microcontroller

Nghia Nguyen, H. 2B-01.5 Establishing Physical Survivability of Large Networks using Properties of Two-Connected Graphs

Ngo, B. 1C-01.1 Using Rate Balancing to Improve Throughput in a Wireless Multi-hop Chain Topology

Nguyen, C. 1D-08.1 Optimizing the FBI Fingerprint Compression Algorithm for BLACKFIN Processor

Nguyen, D. 1C-06.4 Viability of a wavelet-based multicarrier modulation system

Ni, R. 2C-02.3 Quantified Cost-Balanced Overlay Multicast Routing for Multimedia Network

Nishihara, A. 1C-03.5 Subregion Technique with Optimal Local Parameter Sets for Image Deconvolution

Niu, C. 1C-10.2 The enactment of optimal target value in power plant operation based on language association rule

Niu, C. 2B-10.4 The Application of Fuzzy Data Mining in Coal-Fired Boiler Combustion System

Noordin, N. 3D-06.4 A Simplified Sphere Decoding Algorithm for MIMO Transmission System

Noordin, N. 2B-09.1 Full Rate Concatenated Space-Frequency and Space-Time OFDM over Naftali Fading Channels

Noordin, N. 3D-03.4 Applications of Low Voltage FGMOS based CCII in Filtering

Nor Liza Wan Mahadi, W. 3C-08.3 Thermal Analysis of a Tubular Permanent Magnet Linear Generator Using Multiphysics Solver

Nualhong, D. 1D-12.1 Generation Expansion Planning Including Biomass Energy Sources with

Nualhong, D. 1D-12.2 Relative Velocity Updating in Parallel Particle Swarm Optimization Based

Oetomo, D. 3C-04.3 Robust Control Framework for Piezoelectric Actuation Systems in Micro/Nano Manipulation

Oguchi, K. 1C-01.5 Effective Verification of Wavelength Routing Function by Wavelength Transfer matrix Method

Performance Analysis of Noncoherent UWB Transceiver with PEWB under LOS Residential Channel
Oh, H. 2D-04.2
Environment

Oh, J. 1B-11.2 The Detection of Curve-type Defects in the TFT-LCD panels with Machine Vision

Ohyama, M. 1D-02.2 Support Tool Using Positions of Contents for Voice/Gaze-controlled Browsers

Oikawa, S. 2C-12.4 A Parallel Estimator with LMS Adaptation for Fast Fading Channels

Okada, K. 1C-01.5 Effective Verification of Wavelength Routing Function by Wavelength Transfer matrix Method

Okamoto, E. 1C-06.3 A Wavelet Packet Modulation Method for Satellite Communications

Okamoto, E. 2D-04.3 A higher rate turbo equalization with multilevel block-coded modulation

Oliver, T. 1D-10.1 Reconfigurable Computing: Peripheral Power and Area Optimisation Techniques

Omar, M. 3C-10.2 A New Approach for Harmonic Compensation Using Single- phase Hybrid Active Power filter

Ong, K. 1C-08.4 A Study of Power Asymmetry Ratio of Single Trial Electroencephalogram During Finger Movement

Tencon 2005 Page 164

Name and Initial Session Paper Title

Ono, O. 3B-07.4 Direct-Proportional Length-Based DNA Computing Implementation for Elevator Scheduling Problem

Experimental Comparison of Initial Pool Generation for Direct-Proportional Length-Based DNA

Ono, O. 1B-08.3
Computing

Orchard, K. 2C-10.4 Unmanned Light Craft Balance System Using Fuzzy Logic Control

Orchard, K. 3C-04.4 Microcontroller-Based Autonomous Light Craft

Osbrough, S. 1B-02.3 A Heuristic for low power QoS routing in ad hoc networks

O'Shea, P. 3C-08.2 Rapid Detection of Changes to Individual Modes in Multimodal Power Systems

O'Shea, P. 1C-06.2 An algorithm for estimating time-varying Doppler at low SNR

O'Shea, P. 1B-10.4 Damping Estimation In Highly Interconnected Power Systems

O'Shea, P. 3B-03.3 Efficient Structure for Single-Bit Digital Comb Filters and Resonators

O'Shea, P. 2D-07.3 A Single-Bit Digital DC-Blocker Using Ternary Filtering

Ou, L. 2D-11.4 Two-phases Parallel Neural Network Algorithm Based on RPROP

P. Gopalan, R. 2B-03.2 Dynamic Router Tables for Full Expansion/Compression IP Lookup

P. K., L. 3D-03.1 A Fast Transient Response Single Phase Active Power Filter

P. Shirochin, V. 3B-10.1 An intelligent lightweight Intrusion Detection System(IDS)

Pachori, R. 2D-05.2 A Novel Technique for Merging of Multisensor and Defocussed Images using Multiwavelets

Pahlovy, S. 3D-10.3 New method for nano hardness measurement of single crystal Aluminum

Painkras, E. 3D-05.3 A Framework for the Design and Implementation of a Dynamic Face Tracking System

Pal, A. 3D-09.5 Logic Synthesis and Technology Mapping of MUX-based

Pal, K. 3D-03.4 Applications of Low Voltage FGMOS based CCII in Filtering

Palazzi, C. 1D-04.2 High Speed Over Ocean Radio Link to Great Barrier Reef

Palmer, R. 2B-03.4 Fast MPLS Network Optimisation using Machine Learning

Pang, K. 1C-09.1 Performance analysis and comparison of trellis coded STTD and TSTD schemes for CDMA system

Panigrahi, B. 2C-06.1 Intelligent Classification of Power Quality Disturbances in a Distribution Network

Paramesran, R. 3B-10.2 An Image Watermarking Scheme Based on Orthogonal Moments

Paramesran, R. 1C-08.4 A Study of Power Asymmetry Ratio of Single Trial Electroencephalogram During Finger Movement

Park, C. 3B-03.4 Design of DMB Nonuniform SIR-type BPF with Embedded Tunable Pads Using LTCC

Park, C. 2B-08.3 A Study on Constructing the Arithmetic Unit over GF(3m)

Analog Baseband Chain in a 0.18 µm Standard Digital CMOS Technology for IEEE802.15.3a (UWB)
Park, E. 1C-05.4
Receiver

Park, K. 2C-07.3 Adaptive Surface Fitting for Inspection of FPD Devices Using Multilevel B-spline Approximation

Park, K. 1B-11.2 The Detection of Curve-type Defects in the TFT-LCD panels with Machine Vision

Patanukhom, K. 1C-03.5 Subregion Technique with Optimal Local Parameter Sets for Image Deconvolution

Patel, N. 3C-10.4 Accurate Estimation of Harmonic Components of Power Signals

Patel, N. 3B-04.3 Online Implementation of Servo Controllers Using Bit-Streams

Tencon 2005 Page 165

Name and Initial Session Paper Title

Pathak, S. 3D-08.2 Leakage Current Estimation of H.V Insulators on Power Distribution Lines

Pathirana, P. 3D-01.3 Location based power control for Mobile devices in a Cellular Network

Pathirana, P. 3B-04.4 Robust terminal attitude navigation for autonomous vehicles

Pathirana, P. 3B-01.3 Sensor Fusion Based Localization of a Mobile User in a Wireless Network

Pathirana, V. 1B-07.4 A Complete Isothermal Model for the Lateral Insulated Gate Bipolar Transistor on SOI technology

Pawlikowski, K. 3B-09.1 A Methodology for Improving PESQ Accuracy for Chinese Speech

Pei, R. 3D-04.3 Real-Time Monitoring and Integrated Automation Control in Steel Rolling Plant

Peng, D. 3B-04.2 Dynamic Modeling and Controller Designing for Flexible Liquid-filled spacecraft

Peng, G. 2D-03.5 The Design and Implementation of a WSRF-Compliant Information Service on Wireless Grids

Sustainable Agriculture Mobile Commerce (SamCom) Extending the rural grocery supply chain through
Perreau, S. 3C-01.3
M-Commerce

Pervouchine, V. 3C-05.4 Document examiner feature extraction: thinned vs. skeletonised handwriting images

Petchakit, W. 3B-03.1 Electronically current mode all-pass filter with only one grounded capacitor

A Solar-powered Battery Charger with Neural Network Maximum Power Point Tracking Implemented
Petchjatuporn, P. 1B-07.3
on a Low-Cost PIC-microcontroller

Phaiboon, S. 2C-10.2 2 to 16 GHz Microwave Line-of-Sight Path Loss Prediction on Urban street by Fuzzy Logic Models

Robust Position Control of a Synchronous Motor Using Discrete Sliding mode Model Following
Phakamach, P. 3D-08.1
Controller

Pham, D. 2C-03.2 Performance of IEEE 802.11b Wireless Links: An Experimental Study

Phoomvuthisarn, S. 1D-06.1 Minimization of transmission loss with consideration of transformer taps constraint

Phoomvuthisarn, S. 1D-12.2 Relative Velocity Updating in Parallel Particle Swarm Optimization Based

Phophongviwat, T. 1D-06.4 Minimum Cost Design of Small Low-Loss Transformers

Phumiphak, T. 3D-10.4 Energy Management Program for Use of Induction Motors Based on Efficiency Prediction

A Nonparametric Discriminant Approach in Resolving Complex Multi-class Query for Content-based

Ping Chung, K. 1C-03.4
Image Retrieval

Piper, I. 1C-04.2 Introducing NTS: A Classification System for Knowledge Library Applications

Po Wong, K. 1C-02.3 Intelligent Customer Relationship Management On The Web

Pose, R. 3C-02.5 The ARP Virtual Reality System in Addressing Security Threats and Disaster Scenarios

Pose, R. 3D-02.2 Security Models in the Password-Capability System

Pose, R. 1B-02.4 An Intrusion Detection System for Suburban Ad-hoc Networks

Pragtong, P. 3C-01.2 Modeling of VoIP Conversational Speech in Noisy Environment

Prakash, E. 3D-05.1 Face personalization: animated facial modeling approach using radial basis function

Prasad, K. 1C-10.1 A Novel Scheme of Load forecasting pertaining to long term planning of a distribution system

Implementation of Fuzzy Supervisory PI Controllers for a Binary Distillation Column Using Fuzzy c-
Pratishthananda, S. 2B-10.3
Means Clustering Combined with Fuzzy Gain Scheduling

Pratishthananda, S. 2C-10.5 Real-Time Control of a Binary Distillation Column Using HGA Fuzzy Supervisory PI Controllers

The Development of a Novel Image Quality Metric and a Synthetic Colour Test Image for Objective
Punchihewa, A. 1C-03.2
Quality Assessment of Digital Codecs

Pungprasert, V. 1D-12.1 Generation Expansion Planning Including Biomass Energy Sources with

Tencon 2005 Page 166

Name and Initial Session Paper Title
An Approach to Resolve Integer Ambiguity of GPS Carrier Phase Difference for Spacecraft Attitude
Purivigraipong, S. 1D-09.2
Determination

Purivigraipong, S. 1D-09.3 Spacecraft Orbit Determination from Code Information

Qi, D. 1D-05.3 A variable low bit rate Speech coder based on MELP

Qiaoping, S. 2B-07.1 A Geometric Mean Based Adaptive Local Noise Removal Algorithm

Qiu, B. 2C-11.4 Mobility Prediction with Direction Tracking on Dynamic Source Routing

Qiu, B. 1C-01.4 Traffic Measurement for Large Internet Flows Using Multi-stage Landmark LRU Cache

Quang Fang, J. 1B-07.2 Development of a conductive photoresist with a mixture of SU-8 and HCl doped polyaniline

Que, Y. 1D-02.4 Rendering Models for Immersive Voice Communications within Distributed Virtual Environment

Quevedo, D. 2B-08.5 Probability of Interpolation for a Mute Sample Interpolative A/D Converter with Horizon-Length Two

Qumar, J. 2D-05.2 A Novel Technique for Merging of Multisensor and Defocussed Images using Multiwavelets

Qunfei, Z. 2C-05.3 M-ary Chirp Spread Spectrum Modulation for Underwater Acoustic Communication

Ra Kim, M. 1B-10.1 A Design of the Ubiquitous-based Electric Power Control Framework

Raghavan, S. 3B-07.1 Genetic Algorithm for Classification of Epilepsy Risk Levels from EEG Signals

Rahman, A. 1C-03.1 A motion-based approach for temporal texture synthesis

Rai, S. 2B-03.2 Dynamic Router Tables for Full Expansion/Compression IP Lookup

Rajendran, B. 1B-11.5 Establishment of Community Information Network in a Developing Nation

Rajgopal, K. 3C-03.2 A Delayless Adaptive IFIR Filterbank Structure for Active Noise Control

Rajput, S. 3D-03.4 Applications of Low Voltage FGMOS based CCII in Filtering

Rakluea, P. 3B-06.4 Analysis of Right Angle Microstrip Slot Antenna

Ramachandran, N. 2B-10.2 Fuzzy Matching Scheme for Stereo Vision based Electronic Travel Aid

Ramakrishnan, K. 3D-05.2 3D Shape Extraction of Human Face in Presence of Facial Hair: A Profilometric Approach

Ramanan, B. 3D-07.1 Clustering based Video Transcoding for Congestion Control in Wireless Environments

A Survey of FPGAs for Acceleration of High Performance Computing and their Application to
Ramdas, T. 3D-09.2
Computational Molecular Biology

Ramiah, H. 2C-12.7 High Linearity 0.18um CMOS Mixer with Current Draining Technique

Randhawa, S. 1B-03.3 CFA Demosaicking with Improved Colour Edge Preservation

Rangan, K. 2C-12.9 An FPGA-based Architecture for Block Matching Motion Estimation Algorithm

Ranjan Poddar, D. 3B-06.3 Novel Dielectric Rectangular Slab Antenna With Improved Performance

Ranjan, R. 1C-10.1 A Novel Scheme of Load forecasting pertaining to long term planning of a distribution system

Rao Lolla, K. 3D-05.2 3D Shape Extraction of Human Face in Presence of Facial Hair: A Profilometric Approach

Rathgeb, E. 1C-04.4 The Performance of Reliable Server Pooling Systems in Different Server Capacity Scenarios

Ray, B. 3D-06.2 BASEBAND PROCESSOR FOR A GLOBAL PoSITIONING SYSTEM RECEIVER

Rayner, E. 1C-04.2 Introducing NTS: A Classification System for Knowledge Library Applications

Rayudu, R. 2D-12.1 A Distributed Architecture for Power Network Fault Analysis

Tencon 2005 Page 167

Name and Initial Session Paper Title

Reddy Vustikayala, S. 2C-12.9 An FPGA-based Architecture for Block Matching Motion Estimation Algorithm

Reddy, M. 2C-12.9 An FPGA-based Architecture for Block Matching Motion Estimation Algorithm

Redekar, G. 2D-04.4 Equivalence of the Linear Symbol-level MMSE and the Generalized RAKE

Regan, M. 3C-02.5 The ARP Virtual Reality System in Addressing Security Threats and Disaster Scenarios

Rehman, M. 1D-07.3 Development of an Analog Electronic Wattmeter

Rehman,Saad, M. 1D-07.3 Development of an Analog Electronic Wattmeter

Common and Separate Parameterizations of Vision Model Based Perceptual Post Filtering For Digital
Ren Wu, H. 2D-06.3
Colour Images

Ren Wu, H. 1D-08.4 An Improved Visual Pruning Algorithm for Perceptually Lossless Medical Image Coding

Ren Wu, H. 2D-12.9 Visual Assessment for The Quantization Error in Wavelet Based Monochrome Videos

Reza Sayyah Jahromi, M. 2D-07.5 An Efficient Implementation of Generalized Sidelobe Canceler for Broadband Antenna Array Signals

Rhodette Pedrasa, J. 1D-02.5 An Enhanced Framing Strategy for Jitter Management

Richard Hizon, J. 1B-07.1 Spiral Inductor Coupling on 0.25 um Epitaxial CMOS Process

Richard Hizon, J. 1D-07.4 Monolithic Spiral Inductors for a 0.25 um Digital CMOS Process

Richardson, B. 1B-11.3 The Exograsp delivers tactile and kinaesthetic information about virtual objects

Risson, J. 1C-02.1 Towards Robust Internet Applications: Self-Organizing Overlays

Roh, J. 2C-12.5 Dynamic Spectrum Allocation Using Channel Power in CDMA Networks

Rohana Adi, S. 3C-08.3 Thermal Analysis of a Tubular Permanent Magnet Linear Generator Using Multiphysics Solver

Rojas, A. 2B-11.3 Validation of the Random Waypoint Mobility Model Through a Real World Mobility Trace

Romli, M. 2B-02.3 Scalable AWG-based Multi-Protocol Optical Switch (MPOS) for Future Optical GRID Networking

Improved Dynamic Multimedia Resource Adaptation-based Peer-to-Peer Sys-tem through Locality-

Rong, L. 2D-02.3
based Clustering and Service

Rosales, M. 1B-07.1 Spiral Inductor Coupling on 0.25 um Epitaxial CMOS Process

Rosales, M. 1D-07.4 Monolithic Spiral Inductors for a 0.25 um Digital CMOS Process

Roy, S. 3C-09.3 Quasi-Static Analysis of Defected Ground Structure

Ruan, Q. 1B-03.4 Feature-Oriented Coupled Bidirectional Flow for Image Denoising and Edge Sharpening

Ruan, Q. 2C-04.3 Adaptive Multiresolution Representation of a Point-Based 3D Model for Rendering

Rudra, T. 1D-05.4 Cognitive Emotion in Speech Interactive Games

Rudra, T. 2D-12.6 Analysis of Conversational Game Pidgin Language

Ruighaver, A. 3C-02.4 An Information-Centric Approach to Data Security in Organizations

Ruighaver, A. 3D-02.3 Wireless Intrusion Detection: Not as easy as traditional network intrusion detection

Russ, G. 1D-07.1 Detection of Faulty Semiconductor Wafers using Dynamic Growing Self Organizing Map

Sadik, A. 3B-03.3 Efficient Structure for Single-Bit Digital Comb Filters and Resonators

Sadik, A. 2D-07.3 A Single-Bit Digital DC-Blocker Using Ternary Filtering

Sadik, A 2D-12.13 Adaptive LMS Ternary Filtering

Tencon 2005 Page 168

Name and Initial Session Paper Title

Safaei, F. 1D-02.4 Rendering Models for Immersive Voice Communications within Distributed Virtual Environment

Sakakibara, H. 1C-06.3 A Wavelet Packet Modulation Method for Satellite Communications

Sakaue, K. 2B-04.5 Robust Background Subtraction based on Bi-polar Radial Reach Correlation

Sakaue, K. 2C-04.1 Development of Intelligent Wheelchair System based on Stereo Omni-directional System (SOS)

Saliba, A. 3B-01.4 Perceptual Influences on Use of Fixed Lines and Mobile Phones

Salman, M. 2D-07.2 State-Space Least Mean Square with Adaptive Memory

Samanta, D. 2D-08.1 User Model to Design Adaptable Interfaces for Motor-Impaired Users

Sami, A. 1C-08.5 Decision Tree Construction for Genetics Applications Based on Association Rules

Sangswang, A. 2C-12.1 An Experimental Study of Random Noise Characteristics in a Power Converter

Sangtong, S. 1C-07.4 Low-voltage CMOS instrumentation amplifier for piezoresistive transducer

Sankar Kottareddy Gari, J. 1C-06.1 A Wavelet Based Feature Extraction For Voice Lock Systems

Sappok, S. 2B-08.2 Calibration Technique for Sigma-Delta Modulation Loops using Radio-Frequency Phase Detection}

Saravanan, N. 3D-01.4 Design and Performance Evaluation of an Inter-System Handover Algorithm in UMTS/GSM Networks

Sastry, S. 2D-11.6 Algebraic Semantics for Complete Interaction Sequence

Sathananthan, K. 2C-09.2 Low-Complexity Diversity Enhanced OFDM for Wireless Communications

Sato, M. 2D-10.5 A Self-Distributing Software Environment with Binary Interception for Multi-Projector Displays

Sato, M. 2C-02.5 Kansei-based investigation of methods for projection of digital content

Sato, M. 3D-05.5 Human-Machine Interaction Compensating Imperfection in Facial Expression Recognition

Sato, W. 2B-05.3 Topologically Robust Surface Reconstruction for 3D Shape Modification

Satoh, Y. 2B-04.5 Robust Background Subtraction based on Bi-polar Radial Reach Correlation

Satoh, Y. 2C-04.1 Development of Intelligent Wheelchair System based on Stereo Omni-directional System (SOS)

Saufee Muhammad, M. 3B-07.4 Direct-Proportional Length-Based DNA Computing Implementation for Elevator Scheduling Problem

Experimental Comparison of Initial Pool Generation for Direct-Proportional Length-Based DNA

Saufee Muhammad, M. 1B-08.3
Computing

Schmidt, B. 3B-07.3 Using Graphics Hardware to Accelerate Biological Sequence Analysis

Schmidt, B. 1B-08.1 Mapping of Genetic Algorithms for Protein Foilding onto Computational Grids

Schroeder, A. 3B-07.3 Using Graphics Hardware to Accelerate Biological Sequence Analysis

Seachia, S. 1D-09.4 Human Identification System Based ECG Signal

Design of Multicast Routing and Wavelength Assignment in Multifiber WDM Mesh Networks for
Segkhoonthod, S. 2B-03.3
Asymmetric Traffics

Sekercioglu, A. 1D-01.1 Trust Models and Security Considerations in Multicast Listener Discovery Protocol version 2 (MLDv2)

Sekercioglu, A. 2C-03.2 Performance of IEEE 802.11b Wireless Links: An Experimental Study

Senaratne, R. 3C-05.5 Using Particle Swarm Optimisation for Elastic Bunch Graph Matching to Recognise Faces

Sengupta, S. 1B-04.1 Automated Translation of Behavioral Models using OCL and XML

Semantic Indexing of News Video Sequences: A Multimodal Hierarchical approach based on Hidden
Sengupta, S. 3D-07.4
Markov Model

Tencon 2005 Page 169

Name and Initial Session Paper Title

Sentosa Liong, S. 2C-11.5 Bluetooth Network Security: A New Approach to Secure Scatternet Formation

Sethuraman, V. 3C-03.2 A Delayless Adaptive IFIR Filterbank Structure for Active Noise Control

Shadich, R. 2B-08.1 A scalable parallel computational core for embedded processing

Shah Alam, M. 2C-01.2 Block Based Window Retransmission ARQ Scheme for 100Mbit/s Infrared Links

Shahid, M. 2B-04.4 Novel Masks for Multimodality Image Fusion using DTCWT

Shahrestani, S. 2B-11.1 Wireless Ad-Hoc Networking: Analysis and Validation of Simulation Results

Shakowat Zaman Sarker, M. 2B-10.5 A Proposed Air-Conditioning System Using Fuzzy Algorithm for Industrial Application

Shamim Ara Shawkat, M. 2C-01.2 Block Based Window Retransmission ARQ Scheme for 100Mbit/s Infrared Links

Shang, L. 2D-09.4 A Frequency-Selective Mobile Radio Channel with Hyperbolically Distributed Scatterers

Shang, L. 2D-07.4 Spectral Analysis of Delta and Sigma-Delta Modulators Using Linear FM Input Signals

Sharda, N. 2D-02.1 Implementation and Usability of User Interfaces for Quality of Service Management

Sharma, S. 3D-03.4 Applications of Low Voltage FGMOS based CCII in Filtering

Sharmeen, R. 1D-07.2 Frequency and Flexibility Based Cell Phone Keypad Layout

Shekhar Jamuar, S. 3D-06.4 A Simplified Sphere Decoding Algorithm for MIMO Transmission System

Shekhar Jamuar, S. 2B-09.1 Full Rate Concatenated Space-Frequency and Space-Time OFDM over Naftali Fading Channels

Generation of Excitation Signal in Voice excited Linear Predictive Coding using Discrete Cosine
Shekhar, P. 1D-05.1
Transform

Shen, J. 1D-02.3 An Approach for OWL-Based Web Services Components and Composition Framework

Shen, L. 1C-12.5 Output-coding and SVM for multiclass microarray classification

Shenoy, R. 3D-07.5 Improved Algorithms for Implementation of MPEG2 AAC Decoder on FPGA

Shepherd, R. 1D-03.4 Bluetooth Based Proximity Sensing for Reactive Mobile Robots

Shevlyakov, G. 2B-07.2 A Linear Fusion of the Local Kalman Estimates

Shi, Z. 2D-11.5 Prime Implicant Computation Based on Adjacent Minterm Operations

Equalization of Time Variant Multipath Channels Using Channel Estimation Based Classification
Shimamura, T. 2D-09.3
Techniques

Shimamura, T. 2C-12.4 A Parallel Estimator with LMS Adaptation for Fast Fading Channels

Shimizu, T. 2C-02.5 Kansei-based investigation of methods for projection of digital content

Shin, V. 2B-07.2 A Linear Fusion of the Local Kalman Estimates

Shiraishi, S. 3C-03.1 A Steady-State Analysis of a CORDIC-Based Adaptive ARMA Lattice Filter

Shirakawa, I. 1C-03.3 A Sampling Switch Design for Liquid Crystal Displays

Shirinzadeh, B. 2B-12.2 Assembly Modelling Through Constraint-based Manipulations in A Virtual Reality Environment

Shirinzadeh, B. 3C-04.3 Robust Control Framework for Piezoelectric Actuation Systems in Micro/Nano Manipulation

Shukri, Z. 2B-06.2 COMTRADE-Based Fault Information System for TNB Substations

Shun Tan, Y. 1C-08.2 Bluetooth-enabled ECG Monitoring System

Geometrical and Phasor Representation of Multi-level Amplitude-Phase Modulation Formats and

Siang Tiong, H. 1B-05.5
Photonic Transmitter Structures

Tencon 2005 Page 170

Name and Initial Session Paper Title

Sik Choi, S. 2C-12.6 Performance Evaluation of CDMA Systems Based on User Location

Siliquini, J. 2D-01.3 Internet Transfer Capabilities for Time Critical Communications

Simmonds, A. 1B-01.5 PIPPON: A Physical Infrastructure-aware P2P Overlay Network

Singh, M. 1B-04.5 Masking approach to secure systems from Operating system Fingerprinting

Singh, S. 1B-02.3 A Heuristic for low power QoS routing in ad hoc networks

Siong Lim, H. 1B-03.2 Medical Image Processing: From Lossless to Lossy Compression

Sirisantisamrid, K. 2D-06.1 Determine Calibration Parameters with Satisfied Constraints for Coplanar Camera Calibration

Grid-connected Photovoltaic System with Maximum Power Point Tracking using Self-Organizing Fuzzy
Sirisuk, P. 2C-10.3
Logic Controller
A Solar-powered Battery Charger with Neural Network Maximum Power Point Tracking Implemented
Sirisuk, P. 1B-07.3
on a Low-Cost PIC-microcontroller

Sison, L. 1D-12.4 Energy-saving Flyback Converter for EDM Applications

Sithitavorn, K. 2C-11.4 Mobility Prediction with Direction Tracking on Dynamic Source Routing

Siva, S. 3D-05.2 3D Shape Extraction of Human Face in Presence of Facial Hair: A Profilometric Approach

Sivahumaran, T. 1C-09.1 Performance analysis and comparison of trellis coded STTD and TSTD schemes for CDMA system

On the Design of Space-Frequency Codes from Super-Orthogonal Space-Time Trellis Codes via
Siwamogsatham, S. 1C-12.2
Mapping

Slay, J. 2C-03.1 The 802.11 Technology Gap Case Studies in Crime

Smith, A. 1D-07.1 Detection of Faulty Semiconductor Wafers using Dynamic Growing Self Organizing Map

Soh, B. 1C-02.4 Context Based Riskiness Assessment

Soh, S. 2B-03.2 Dynamic Router Tables for Full Expansion/Compression IP Lookup

Sohel, F. 3D-10.1 A Novel Half-Way Shifting Bezier Curve Model

Implementation of Fuzzy Supervisory PI Controllers for a Binary Distillation Column Using Fuzzy c-
Somsung, K. 2B-10.3
Means Clustering Combined with Fuzzy Gain Scheduling

Song, H. 2B-02.5 Cost Model based Configuration Management Policy in OBS Networks

Common and Separate Parameterizations of Vision Model Based Perceptual Post Filtering For Digital
Soon Tan, C. 2D-06.3
Colour Images

Soon Tan, C. 1D-08.2 Subjective Image Quality Assessment at the Threshold Level

Sreenivasulu, N. 3D-01.4 Design and Performance Evaluation of an Inter-System Handover Algorithm in UMTS/GSM Networks

Design and Performance Evaluation of Channel Reservation Schemes for QoS Constrained Multimedia
Srichavengsup, W. 2D-06.5
Traffic
A Novel Wavelet Packet Transform Based Feature Extraction Algorithm For Image Texture
Srinivasan, R. 2C-09.4
Classification
A Novel Wavelet Packet Transform Based Feature Extraction Algorithm For Image Texture
Srinivasan, S. 2C-09.4
Classification

Stewart, L. 2C-11.1 An Architecture for Automated Network Control of QoS over Consumer Broadband Links

Stootman, F. 2B-11.1 Wireless Ad-Hoc Networking: Analysis and Validation of Simulation Results

Su Chung, Y. 2C-07.3 Adaptive Surface Fitting for Inspection of FPD Devices Using Multilevel B-spline Approximation

Su Chung, Y. 1B-11.2 The Detection of Curve-type Defects in the TFT-LCD panels with Machine Vision

Su, J. 3B-10.1 An intelligent lightweight Intrusion Detection System(IDS)

Evaluating the Restorability in WDM Networks Based on Directional Cycle Decomposition Algorithm
Sue, C. 1B-01.3
(DCDA)

Tencon 2005 Page 171

Name and Initial Session Paper Title

Sue, C. 1B-05.4 A Robust Design of Reconfigurable Coder/Decoder Against Un-authorized Users

Sugano, Y. 1D-09.5 Fast and Accurate Positioning Technique Using Ultrasonic Phase Accordance Method

Sugawara, H. 3B-06.1 A Reconfigurable RF Circuit Architecture for Dynamic Power Reduction

Sugimoto, M. 1D-09.5 Fast and Accurate Positioning Technique Using Ultrasonic Phase Accordance Method

Sugiyama, Y. 2B-04.3 A Similarity Analysis of DGGE Images using DP matching

Sukanesh, R. 3B-07.1 Genetic Algorithm for Classification of Epilepsy Risk Levels from EEG Signals

Sun, Y. 3B-02.1 Merging Retrieval Results in Hybrid Peer-to-Peer Networks

Sun, Y. 2C-02.2 A Novel Grid-based Content Management System

Suraweera, H. 1B-12.5 Error Performance Analysis of Clipped Alamouti Space-Time Coded OFDM Systems

Suraweera, H. 1C-12.1 Alamouti Coded OFDM in Rayleigh Fast Fading Channels - Receiver Performance Analysis

Suryasaputra, R. 2D-01.4 OptiFlow - A Capacity Management Tool

Suter, D. 1B-08.4 A Semi-automatic System for Measuring Tibial Cartilage Volume

Suyono, H. 2B-12.9 Development of Power System Analysis Software Using Object Components

Suyono, H. 2B-05.2 Transient Stability Program Using Component-Based Software Engineering

Identification of Nonlinear Differential Equation Models from Generalized Frequency Response

Swain, A. 3B-03.2
Functions

Swain, A. 3C-10.4 Accurate Estimation of Harmonic Components of Power Signals

Swain, A. 3B-04.3 Online Implementation of Servo Controllers Using Bit-Streams

Sustainable Agriculture Mobile Commerce (SamCom) Extending the rural grocery supply chain through
Swatman, P. 3C-01.3
M-Commerce

Swiegers, G. 3C-09.3 Quasi-Static Analysis of Defected Ground Structure

Symmons, M. 1B-11.3 The Exograsp delivers tactile and kinaesthetic information about virtual objects

T.-C. Chen, O. 3D-07.2 Design of Limited Buffer Size for Storing Multiple Reference Frames in a Video Codec

Tahtali, M. 2C-04.5 Restoration of Non-Uniformly Warped Images Using a Typical Frame as Prototype

Takahashi, K. 2D-08.5 Macroaction Synthesis for Agent Systems

Takahashi, K. 3B-05.4 A Study on Face Similarity Using Knowledge Processing Methods

Takahashi, M. 1C-08.5 Decision Tree Construction for Genetics Applications Based on Association Rules

Takahashi, S. 1C-03.3 A Sampling Switch Design for Liquid Crystal Displays

Takasaki, Y. 1B-05.1 Versatile Multiplexing Based on Hunting-Free Framing

Takeda, N. 3B-05.4 A Study on Face Similarity Using Knowledge Processing Methods

Tan, C. 1B-02.2 Modifying AODV for Efficient Power-Aware Routing in MANETs

Tan, D. 1D-08.4 An Improved Visual Pruning Algorithm for Perceptually Lossless Medical Image Coding

Tan, D. 1D-08.2 Subjective Image Quality Assessment at the Threshold Level

Tan, L. 3B-02.4 On Congestion Control in Multicast Networks

Tan, L. 2D-03.3 Price-based Max-Min Fair Rate Allocation in Wireless Multi-hop Networks

Tencon 2005 Page 172

Name and Initial Session Paper Title

Tanaka, M. 2C-05.2 Implementation and Evaluation of the Extended Cell Splitting Algorithm

Tandel, K. 1B-11.5 Establishment of Community Information Network in a Developing Nation

Tang, L. 2C-10.4 Unmanned Light Craft Balance System Using Fuzzy Logic Control

Tang, L. 3C-04.4 Microcontroller-Based Autonomous Light Craft

Tangsangiumvisai, N. 2B-07.4 Improved Performance for Stereophonic Acoustic Echo Cancellation using Subband Structure

Tao, Z. 3D-02.3 Wireless Intrusion Detection: Not as easy as traditional network intrusion detection

Teng, J. 2B-06.4 Embedded Web Server Based Power Quality Recorder

Teng, J. 1D-12.3 Assessments for the Impacts and Benefits of Wind Farm Placement

Thakur, M. 2B-11.8 Link Layer Mitigation in Rural UHF-MIMO Linking Systems

Thanachayanont, A. 1C-07.4 Low-voltage CMOS instrumentation amplifier for piezoresistive transducer

Thanachayanont, C. 1C-07.3 Diode fabricated by layer by layer deposition of semiconductor nanoparticles

Thanachayanont, C. 1C-07.4 Low-voltage CMOS instrumentation amplifier for piezoresistive transducer

Thawani, A. 3D-07.1 Clustering based Video Transcoding for Congestion Control in Wireless Environments

Theresa Gusad, M. 1D-11.1 Design Methodology for CMOS Low-Noise Amplifiers Using Power Matching Techniques

Tian, H. 1B-02.1 CLA-QOS: A Cross-Layer QoS Provisioning Approach for Mobile Ad-hoc Networks

Tick Hui, O. 1B-03.2 Medical Image Processing: From Lossless to Lossy Compression

Detection Schemes for Distributed Space-Time Block Coding in Time-Varying Wireless Cooperative
Ting Cheng, H. 1B-12.3
Systems

Tiraseth, K. 2D-06.1 Determine Calibration Parameters with Satisfied Constraints for Coplanar Camera Calibration

Tongkulboriboon, S. 3B-03.1 Electronically current mode all-pass filter with only one grounded capacitor

Tootell, H. 3D-02.4 Investigating the relationship between Location-Based Services and National Security

Tortal, E. 2D-08.2 Monitoring of the Coalesced Ice Ball in Cryosurgery

Tran, A. 2C-11.3 Ethernet PON or WDM PON: A Comparison of Cost and Reliability

Geometrical and Phasor Representation of Multi-level Amplitude-Phase Modulation Formats and

Tran, D. 1B-05.5
Photonic Transmitter Structures
Long-haul ASK and DPSK Optical Fibre Transmission Systems: Simulink Modeling and Experimental
Tran, D. 1C-05.2
Demonstration Test-Beds

Tran, M. 2B-05.4 End-users resource consumption of spam and a 3D anti-spam evaluation framework

An Approach to FPGA-based Time-Frequency Spectrogram by Real-Time Sweep Spectral Extraction

Tsai, G. 2D-05.1
Algorithm

Tsai, P. 2D-03.4 Extending Cluster File Systems beyond Last Miles

Experimental Comparison of Initial Pool Generation for Direct-Proportional Length-Based DNA

Tsuboi, Y. 1B-08.3
Computing
Equalization of Time Variant Multipath Channels Using Channel Estimation Based Classification
Tsuda, Y. 2D-09.3
Techniques

Tsuda, Y. 2C-12.4 A Parallel Estimator with LMS Adaptation for Fast Fading Channels

Tsukiyama, S. 1C-03.3 A Sampling Switch Design for Liquid Crystal Displays

Equalization of Time Variant Multipath Channels Using Channel Estimation Based Classification
Tsutsumi, Y. 2D-09.3
Techniques

Tu, X. 1C-02.5 The Design of China Tourism E-Business Platform Based on J2EE Technology

Tencon 2005 Page 173

Name and Initial Session Paper Title

Tuan Nguyen, Q. 1C-06.4 Viability of a wavelet-based multicarrier modulation system

Tucker, R. 2C-11.3 Ethernet PON or WDM PON: A Comparison of Cost and Reliability

Tung, H. 2D-03.4 Extending Cluster File Systems beyond Last Miles

Turker, E. 3D-04.4 Magnetic Bearing Controller Tuning Utilizing Server/Client Technology

Turnbull, B. 2C-03.1 The 802.11 Technology Gap Case Studies in Crime

Tuwanut, P. 1C-07.5 A Novel Monostable Multivibrator Circuit

Tyan, L. 3C-04.2 A Resilient Guaranteed Cost Control Design Via the Robust Two-Stage LQ Reliable Control

Ueda, H. 2D-08.5 Macroaction Synthesis for Agent Systems

Ueda, H. 3B-05.4 A Study on Face Similarity Using Knowledge Processing Methods

Usman, M. 3C-01.1 Performance Analysis of Adaptive Source Rate Control Algorithm (ASRC) for VoIP

V., K. 1B-09.5 Downlink Performance of a Finite Word Length Implemented Rake Receiver for 3G WCDMA

Vahidi, B. 3C-08.1 Model of corona for transient study

Varadarajan, S. 2D-05.4 Prominent Region Based Object Recognition using Complex Wavelets

Varadarajan, S. 3D-07.1 Clustering based Video Transcoding for Congestion Control in Wireless Environments

Varadarajan, S. 2B-11.9 Hybrid P2P Based Self Organising Grids for Efficient Resource Distribution

Varmah, K. 2D-12.11 Importance of Accuracy For Steady State Performance Analysis of 3f Induction Motor

Veeravalli, B. 2D-02.4 Design and Implementation of Stream Partitioning Strategies in a Distributed Multimedia System

Venkat Sreenivas, T. 1D-11.2 A Mixed-Split Scheme for 2-D DPCM Based LSF Quantization

An Adaptive High Resolution Measuring System for Characterizing Transformers Subjected to

Venkataraman, J. 1D-06.5
Broadband Excitation

Verma, A. 3C-06.4 High Gain Compact Circularly Polarized Microstrip Antenna

Viet Phung, Q. 2B-01.5 Establishing Physical Survivability of Large Networks using Properties of Two-Connected Graphs

Viinikainen, A. 2C-01.3 Optimized Pricing and Closed Form Algorithm for WFQ Scheduling

Viinikainen, A. 2D-04.1 A new Modulation Scheme for GSM EDGE

A Game Theoretic Framework for Call Admission Control in CDMA Mobile Multimedia Systems with
Virapanicharoen, J. 1B-01.2
Traffic Asymmetry between Uplink and Downlink

Voss, G. 3B-07.3 Using Graphics Hardware to Accelerate Biological Sequence Analysis

Vuori, J. 2D-04.1 A new Modulation Scheme for GSM EDGE

Wan Kim, T. 2D-08.3 Design and Implement the Stimuli Capsule at In-vitro Experiment

Wang, B. 2D-11.5 Prime Implicant Computation Based on Adjacent Minterm Operations

Wang, C. 2B-06.1 Bayes Method of Power Quality Disturbance Classification

Compression of Power Quality Disturbance Data Based on Energy Threshold And Adaptive Arithmetic
Wang, C. 2C-06.3
Encoding

Wang, C. 2C-04.1 Development of Intelligent Wheelchair System based on Stereo Omni-directional System (SOS)

Wang, D. 1B-05.4 A Robust Design of Reconfigurable Coder/Decoder Against Un-authorized Users

Wang, J. 2B-06.1 Bayes Method of Power Quality Disturbance Classification

Tencon 2005 Page 174

Name and Initial Session Paper Title
Compression of Power Quality Disturbance Data Based on Energy Threshold And Adaptive Arithmetic
Wang, J. 2C-06.3
Encoding

Wang, J. 2B-11.4 A hybrid admission control scheme for wireless networks

Wang, J. 1D-05.2 VLSI Architecture Design for Concatenative Speech Synthesizer

Wang, L. 3D-04.3 Real-Time Monitoring and Integrated Automation Control in Steel Rolling Plant

Wang, Q. 1D-08.5 A New Automatic Gait Recognition method based on the Perceptual Curve

Wang, Q. 3C-06.1 A Hybrid Method for the Calculation of Radiation Field from Opening in Shielding Enclosure

Wang, Q. 2D-05.5 A Novel Approach to Interest Point Detection Based on Phase Congruency

Wang, Q. 3D-04.5 Robust Stability Analysis of Delayed Neural Networks with Polytopic Type Uncertainties

Wang, T. 2C-03.3 An optimized DCF schema with Downlink and Uplink bandwidth control in 802.11 WLAN

Wang, T. 2D-02.2 Investigation on the Content Popularity Distribution under K-Transformation in Streaming Applications

Wang, W. 2B-03.1 Improved Vector Route Algorithm Bases on Baster Topological Space Model

Wang, W. 1B-03.4 Feature-Oriented Coupled Bidirectional Flow for Image Denoising and Edge Sharpening

Wang, X. 1B-04.2 A Dual-Spiral Reengineering Model for Legacy System

Wang, Z. 1D-06.2 Research on Novel Power Transformer Protection Based on Self-Correction Function

Wang, Z. 1D-06.3 A Novel Inductance Calculation Method in Power Transformer Model Based on Magnetic Circuit

Wang, Z. 1B-10.5 A Novel Method on Substation Reactive Power Control with the Fuzzy Logic Method

Wardkein, P. 1C-07.5 A Novel Monostable Multivibrator Circuit

Wardkein, P. 1D-09.4 Human Identification System Based ECG Signal

Watcharabutsarakham, S. 2D-11.7 Spell Checker for Thai Document

Design of Multicast Routing and Wavelength Assignment in Multifiber WDM Mesh Networks for
Wattanavarakul, W. 2B-03.3
Asymmetric Traffics

Webber, D. 1B-11.1 Intellectual Property - Challenges for the Future

Wei Ho, T. 2D-12.7 Compression and Enhancement of Digital Mammograms with Wavelet Footprints

Wen, J. 2C-09.1 A new interpolation method for channel estimation of OFDM systems

Wen, J. 1C-11.1 Basket Bloom Filters for Membership Queries

Wen, X. 3B-02.1 Merging Retrieval Results in Hybrid Peer-to-Peer Networks

Wen, X. 2C-02.2 A Novel Grid-based Content Management System

Wen, Y. 2C-03.5 Path-Based Minimum Power Broadcast Problem in Wireless Networks

White, C. 1D-08.2 Subjective Image Quality Assessment at the Threshold Level

Wickramasinghe, L. 3C-05.2 Enhancing Episodic Associative Memory Models for Practical Use

Wilson, S. 1C-11.3 Characterisation of Combat Identification Technologies

Wiltshire, R. 3C-08.2 Rapid Detection of Changes to Individual Modes in Multimodal Power Systems

Winkler, S. 1D-01.2 AutoMAV - Micro Aerial Vehicles for Airport Surveillance

Winkler, S. 1D-01.3 Multiple Sensor Fusion for Autonomous Mini and Micro Aerial Vehicle Navigation

Tencon 2005 Page 175

Name and Initial Session Paper Title

Wishart, A. 2C-10.4 Unmanned Light Craft Balance System Using Fuzzy Logic Control

Wishart, A. 3C-04.4 Microcontroller-Based Autonomous Light Craft

Wisitsora-At, A. 2D-12.4 Ozone Sensors based on Layered SAW Devices with: InOx / SiNx / 36º YX LiTaO3 Structure

Wlodarski, W. 2D-12.4 Ozone Sensors based on Layered SAW Devices with: InOx / SiNx / 36º YX LiTaO3 Structure

Won, C. 2D-08.3 Design and Implement the Stimuli Capsule at In-vitro Experiment

Wong, K. 1C-02.3 Intelligent Customer Relationship Management On The Web

Wong, K. 1B-11.4 Neural Network Ensembles Based Approach for Mineral Prospectivity Prediction

Woo, S. 2D-08.3 Design and Implement the Stimuli Capsule at In-vitro Experiment

Woods, G. 1D-04.2 High Speed Over Ocean Radio Link to Great Barrier Reef

Woods, G. 1B-05.3 Improving Group Delay Measurement Accuracy using the FM Envelope Delay Technique

Wook Kim, D. 1B-01.1 An Effective Hotspot Cell Management Scheme Using Adaptive Handover Time in 4G Mobile Networks

On the Design of Space-Frequency Codes from Super-Orthogonal Space-Time Trellis Codes via
Woradit, K. 1C-12.2
Mapping

Worapradya, K. 2C-10.5 Real-Time Control of a Binary Distillation Column Using HGA Fuzzy Supervisory PI Controllers

Wu, C. 1D-08.5 A New Automatic Gait Recognition method based on the Perceptual Curve

Wu, C. 2D-05.5 A Novel Approach to Interest Point Detection Based on Phase Congruency

Wu, D. 1D-08.4 An Improved Visual Pruning Algorithm for Perceptually Lossless Medical Image Coding

Wu, D. 1D-08.2 Subjective Image Quality Assessment at the Threshold Level

Wu, J. 3B-06.5 ZWOS A Lightweight Operating System For Lightweight Embedded Wireless Devices

Wu, J. 2C-01.5 Connectionism-Based CBR Method for Distribution Short-Term Nodal Load Forecasting

Wuillemin, D. 1B-11.3 The Exograsp delivers tactile and kinaesthetic information about virtual objects

Wunderlich, R. 2B-08.2 Calibration Technique for Sigma-Delta Modulation Loops using Radio-Frequency Phase Detection}

On the Design of Space-Frequency Codes from Super-Orthogonal Space-Time Trellis Codes via
Wuttisittikulkij, L. 1C-12.2
Mapping
Design of Multicast Routing and Wavelength Assignment in Multifiber WDM Mesh Networks for
Wuttisittikulkij, L. 2B-03.3
Asymmetric Traffics
Design and Performance Evaluation of Channel Reservation Schemes for QoS Constrained Multimedia
Wuttisittikulkij, L. 2D-06.5
Traffic

Wydrowski, B. 2D-03.1 Evaluation of FAST TCP in Low-Speed DOCSIS-based Access Network

Xia, Y. 1B-03.5 Semi-supervised Segmentation of Textured Images by Using Coupled MRF Model

Xiao, W. 1B-02.1 CLA-QOS: A Cross-Layer QoS Provisioning Approach for Mobile Ad-hoc Networks

Xiao, X. 1C-02.3 Intelligent Customer Relationship Management On The Web

Xiao, Y. 2D-11.12 The Soft Encrypting Channel Based on Turbo-code En-decoders for Wireless Data Transmission

Xiao-Jun, H. 3D-08.3 Three-phase fly-back AC/DC converter with novel resonant snubber circuit

Xiao-Lai, Y. 3D-08.3 Three-phase fly-back AC/DC converter with novel resonant snubber circuit

Xiaoming, Z. 2B-07.1 A Geometric Mean Based Adaptive Local Noise Removal Algorithm

Xie, G. 1C-11.1 Basket Bloom Filters for Membership Queries

Tencon 2005 Page 176

Name and Initial Session Paper Title

Xie, K. 1C-11.1 Basket Bloom Filters for Membership Queries

Xie, Y. 2D-11.12 The Soft Encrypting Channel Based on Turbo-code En-decoders for Wireless Data Transmission

Xing Zheng, W. 3D-04.5 Robust Stability Analysis of Delayed Neural Networks with Polytopic Type Uncertainties

Xing Zheng, W. 1C-11.2 An LMI Approach to Exponential Stability Analysis of Neural Networks with Time-Varying Delay

Xing, J. 2D-12.3 Boiler combustion control using the intelligent controllers

Xu, Y. 1D-06.2 Research on Novel Power Transformer Protection Based on Self-Correction Function

Xu, Y. 1D-06.3 A Novel Inductance Calculation Method in Power Transformer Model Based on Magnetic Circuit

Xu, Y. 1B-10.5 A Novel Method on Substation Reactive Power Control with the Fuzzy Logic Method

Xue, Q. 3B-01.1 A New Quasi-Lumped Resonant Cell Concept in Modern RF and Microwave Components

Xue, W. 3D-01.1 A Novel Joint Power and Rate Control Algorithm in Cellular Radio Systems

Xue, Y. 1B-06.1 Application of Wavelet Array Denoising to ESPRIT Algorithm

General Calculation Formula on the Influence of Extraction Steam Pressure Loss on Thermal Economy
Xuelei, Z. 2D-09.5
of Unit

Xuelei, Z. 3D-10.2 The Influence of IGV and Relative Humidity of Air on Performance Map of Axial Flow Compressor

Yaccob, S. 2B-10.2 Fuzzy Matching Scheme for Stereo Vision based Electronic Travel Aid

Yagnik, J. 3D-05.2 3D Shape Extraction of Human Face in Presence of Facial Hair: A Profilometric Approach

Yamada, C. 1D-10.2 An Efficient Temporal Formula Specification Method for Asynchronous Concurrent Systems

Yamaguchi, I. 2B-04.2 A Mobile Motion Capture System Employing Image Transfer

Yamamoto, H. 2C-02.4 Multisource Receiver-driven Layered Multicast

Yamane, S. 2C-02.5 Kansei-based investigation of methods for projection of digital content

Analysis of Nonlinear Propagation and Wave Mixing of Picosecond Pulses in Semiconductor Optical
Yamayoshi, Y. 1C-07.1
Amplifiers

Yan Hui, K. 3C-06.3 A Patch Loaded Dielectric Resonator Antenna

Use of High-Tc Superconducting Tape RF Coils for Non-invasive Fiber Tracking on Diffusion Tensor
Yang, H. 1C-07.2
MRI application

Yang, H. 3C-09.2 Implementation of Bi-2223 HTS tapes as RF Coils for 3T MRI System

Yang, J. 2D-03.5 The Design and Implementation of a WSRF-Compliant Information Service on Wireless Grids

Yang, M. 1D-01.5 An Unconditionally Secure Multi-round Revocation Scheme Using Secret Sharing

Yang, S. 3C-04.2 A Resilient Guaranteed Cost Control Design Via the Robust Two-Stage LQ Reliable Control

Yang, X. 3B-02.2 Quantitative Availability Prediction for a HA Messaging System

Yang, X. 1B-04.2 A Dual-Spiral Reengineering Model for Legacy System

Yang, Y. 3C-08.5 A Self-contained Direct Methanol Fuel Cell with Surface Tension Fuel Delivery

Yao, H. 2C-12.10 Dynamically Authorized Role-Based Access Control for Grid Applications

Yap, E. 2B-12.10 Applications of FACTS Controller for Improving Power Transmission Capability

Yap, P. 3B-10.2 An Image Watermarking Scheme Based on Orthogonal Moments

Yarali, A. 2B-11.2 Internet Protocol Television (IPTV)

Tencon 2005 Page 177

Name and Initial Session Paper Title

Yashiro, K. 1C-09.3 Use of Turbo Coding for Embedded Transmission Scheme in Optical CDMA

Yatani, K. 1D-09.5 Fast and Accurate Positioning Technique Using Ultrasonic Phase Accordance Method

Yee Pang, S. 1B-03.2 Medical Image Processing: From Lossless to Lossy Compression

Yeh, C. 2D-04.5 Bit-interleaving in fast FH/SSMA systems with convolutional codes and turbo codes

Evaluating the Restorability in WDM Networks Based on Directional Cycle Decomposition Algorithm
Yeh, J. 1B-01.3
(DCDA)

Yin Yum, T. 3B-01.1 A New Quasi-Lumped Resonant Cell Concept in Modern RF and Microwave Components

Yin, Y. 1C-02.5 The Design of China Tourism E-Business Platform Based on J2EE Technology

Yin, Z. 2D-02.5 DHT Based Collaborative Multimedia Streaming and Caching Service

Yinghua, M. 1C-11.1 Basket Bloom Filters for Membership Queries

Yixin, Y. 3D-08.4 PDSR in Phase Angle Space and SPM Based Security Monitoring of Power Systems

Yixin, Y. 1C-10.4 A Real-Time Monitor Framework for Static Voltage Stability of Power System

Yixin, Y. 2C-01.5 Connectionism-Based CBR Method for Distribution Short-Term Nodal Load Forecasting

Yokoi, H. 1C-08.1 FES as Biofeedback for an EMG Controlled Prosthetic Hand

Yoon, K. 2D-08.3 Design and Implement the Stimuli Capsule at In-vitro Experiment

Yoshii, S. 3D-06.1 Spectrum Management of DSL Systems in Japan

Youn, C. 2B-02.5 Cost Model based Configuration Management Policy in OBS Networks

Youn, C. 3B-01.5 Wireless Grid Middleware Architecture for Providing Reliable Services

Young Chung, M. 2C-11.9 Performance Evaluation of an Enhanced GDCF

Young Lee, K. 2C-11.6 The Implementation of Security Message Protocol for PDA PUSH Service

Young Lee, K. 2B-12.3 TFRC Fairness Improving Using the Dynamic Loss Rate Measurement

Yu, C. 1D-12.3 Assessments for the Impacts and Benefits of Wind Farm Placement

Yu, J. 2D-02.2 Investigation on the Content Popularity Distribution under K-Transformation in Streaming Applications

Yu, W. 1C-08.1 FES as Biofeedback for an EMG Controlled Prosthetic Hand

Yu, X. 1C-10.2 The enactment of optimal target value in power plant operation based on language association rule

Yu, X. 2B-10.4 The Application of Fuzzy Data Mining in Coal-Fired Boiler Combustion System

Yuasa, M. 1D-02.2 Support Tool Using Positions of Contents for Voice/Gaze-controlled Browsers

Yu-Shen, S. 1D-10.5 An Artificial Neural-Network-Based Approach to Software Reliability Assessment

Yusof Ibrahim, M. 1B-09.1 Optimization of UTRA SHO Algorithm

Yusof Ibrahim, M. 1B-09.4 Downlink Soft Handover Mechanism Performance in WCDMA System

Yusof, S. 2B-05.2 Transient Stability Program Using Component-Based Software Engineering

Computer Aided-Program for Validation of Measuring System from Unit Step Response by Time
Yutthagowith, P. 2C-07.4
Convolution Method

Yuu Seng, L. 1B-12.2 Space-Time Encoded Secure Chaos Communications with Transmit Beamforming

Yuu Seng, L. 1C-09.2 Blind Adaptive Multiuser Detection for Chaos CDMA Communication

Tencon 2005 Page 178

Name and Initial Session Paper Title

Yuyi, S. 3D-05.5 Human-Machine Interaction Compensating Imperfection in Facial Expression Recognition

Double-Pass Discrete Raman Amplifier with FBG for Pump Power Optimization and Improved
Zamzuri Abdul Kadir, M. 2C-12.3
Performance

Zander, S. 2C-11.1 An Architecture for Automated Network Control of QoS over Consumer Broadband Links

Zhang, D. 1C-11.1 Basket Bloom Filters for Membership Queries

Zhang, J. 3C-08.4 Choice of Reference Conductor in Three Phase Systems A Paradigm Shift

Zhang, J. 2B-06.5 Three Phase Generalised Power Theory for the Measurement of Powers in an Electric Power System

Zhang, J. 1B-12.4 Performance Analysis of Multiple-Relay Decode-and-Forward Cooperation System

Zhang, J. 2C-08.4 A Novel Maximum Likelihood Frequency Synchronizer for OFDM Systems

Zhang, J. 3D-04.2 Greedy exploration policy of Q-learning based on state balance

Zhang, L. 3B-04.1 Robust Nonlinear Coordinated Control of TCSC and Generator Excitation

Zhang, L. 3D-04.3 Real-Time Monitoring and Integrated Automation Control in Steel Rolling Plant

Zhang, Q. 3B-02.1 Merging Retrieval Results in Hybrid Peer-to-Peer Networks

Zhang, Q. 2C-02.2 A Novel Grid-based Content Management System

Zhang, X. 2D-03.3 Price-based Max-Min Fair Rate Allocation in Wireless Multi-hop Networks

Zhang, Y. 2D-11.4 Two-phases Parallel Neural Network Algorithm Based on RPROP

Zhang, Z. 3D-04.2 Greedy exploration policy of Q-learning based on state balance

Zhao, H. 1D-02.3 An Approach for OWL-Based Web Services Components and Composition Framework

Zhao, J. 1C-10.4 A Real-Time Monitor Framework for Static Voltage Stability of Power System

Zhao, L. 3C-10.4 Accurate Estimation of Harmonic Components of Power Signals

Zhao, L. 2C-02.4 Multisource Receiver-driven Layered Multicast

Zhao, M. 3B-06.2 MAEC: A Movement-Assisted Energy Conserving Method in Event Driven Wireless Sensor Networks

Zhao, R. 1D-08.5 A New Automatic Gait Recognition method based on the Perceptual Curve

Zhao, R. 1B-03.5 Semi-supervised Segmentation of Textured Images by Using Coupled MRF Model

An Email Classification Scheme Based on Decision-Theoretic Rough Set Theory and Analysis of Email
Zhao, W. 3C-02.1
Security

Zhao, Y. 2D-11.12 The Soft Encrypting Channel Based on Turbo-code En-decoders for Wireless Data Transmission

Zheng, Y. 3D-04.2 Greedy exploration policy of Q-learning based on state balance

Zhengbing, H. 3B-10.1 An intelligent lightweight Intrusion Detection System(IDS)

Zhong, Y. 2B-12.2 Assembly Modelling Through Constraint-based Manipulations in A Virtual Reality Environment

Zhongyang, X. 2D-11.4 Two-phases Parallel Neural Network Algorithm Based on RPROP

Zhou, J. 2C-04.3 Adaptive Multiresolution Representation of a Point-Based 3D Model for Rendering

Zhou, R. 1B-04.4 A BDL-Based E-Commerce Transaction Architecture

Zhou, R. 2D-11.6 Algebraic Semantics for Complete Interaction Sequence

Zhu, L. 3C-06.1 A Hybrid Method for the Calculation of Radiation Field from Opening in Shielding Enclosure

Tencon 2005 Page 179

Name and Initial Session Paper Title
An Email Classification Scheme Based on Decision-Theoretic Rough Set Theory and Analysis of Email
Zhu, Y. 3C-02.1
Security

Zongkai, Y. 3D-09.4 High speed and low area cost FPGA based pattern matching

Zongkai, Y. 2D-02.2 Investigation on the Content Popularity Distribution under K-Transformation in Streaming Applications

Zukerman, M. 3B-02.4 On Congestion Control in Multicast Networks

Zukerman, M. 2D-03.3 Price-based Max-Min Fair Rate Allocation in Wireless Multi-hop Networks

Tencon 2005 Page 180

Acknowledgments
We thank the Faculty of ICT, Swinburne University of Technology and the School of Electrical and
Computer Engineering, RMIT University for their support with meeting rooms.

A special thanks to Irene Thavarajah, Helen McLean and Olga Lipkin from the Events
Management Office at Monash University for their assistance with the conference organisation.
http://www.monash.edu.au/cmo/

Many thanks to all our sponsors and all participants, Student Volunteers for their assistance

Swinburne Press

Tencon 2005 Page 181