(IJCNS) International Journal of Computer and Network Security, 1

Vol. 2, No. 3, March 2010

Cryptanalysis of an efficient biometrics-based

remote user authentication scheme using smart cards
Fuw-Yi Yang 1 and Jian-Sen Wang 2
1
Department of Computer Science and Information Engineering,
Chaoyang University of Technology,
168 Jifong E. Rd., Wufong Township Taichung County, 41349,
Taiwan, R.O.C.
[email protected]
2
Department of Computer Science and Information Engineering,
Chaoyang University of Technology,
168 Jifong E. Rd., Wufong Township Taichung County, 41349,
Taiwan, R.O.C.
[email protected]

2, we give a brief review of Li and Hwang’s proposed

Abstract: The authors Li and Hwang have proposed an
efficient biometrics-based remote user authentication scheme scheme, and then, in Section 3, we demonstrate this
using smart cards. Security is provided through one-way hash scheme’s weaknesses. Finally, we conclude this paper in
functions and biometrics verification. This scheme is more Section 4.
efficient than other related schemes and enables the users to
change their passwords freely. However, there are some flaws in 2. Review of Li and Hwang’s proposed scheme
it, such as vulnerability to impersonation attacks, password-
guessing attacks, and power analysis attacks. Thus, this paper Li and Hwang proposed an efficient biometrics-based
shows that the scheme proposed by Li and Hwang can be remote user authentication scheme using smart cards. The
susceptible to certain types of attacks. scheme is divided into three phases: registration phase,
Keywords: Biometrics, remote user authentication, login phase, and authentication phase. Here, we briefly
impersonation attacks, password-guessing attacks, power analysis introduce the three phases. In Table 1, we list the notations
attacks and abbreviations used in their scheme. The three phases are
as follows:
1. Introduction Table 1: Notations used in their scheme
In an insecure network environment, user authentication Client
Ci
is a significant component of security. Remote user
authentication schemes are used to verify the validity of the Si Server
user login request. In 1981, Lamport [1] proposed a remote Ri Trust registration center
user authentication scheme with verification tables. IDi Identity of the user
However, Hwang and Li pointed out that if the verification
tables were modified or stolen, the remote authentication PWi Password of the user between Ci and Si
system would be influenced. Therefore, in 2000, Li and Bi Biometrics template of the user
Hwang proposed [2] a remote user authentication scheme h (.) One-way hash function
without any verification tables, using smart cards.
Xs Secret information maintained by the Si
In general, ID-based remote user authentication schemes
are based on passwords [3], [4]. As simple passwords are Rc Random number chosen by the Ci
easy to break, many schemes have been proposed to enhance Rs Random number chosen by the Si
the security of the remote user authentication. But the
passwords can be lost, forgotten, or shared with other || Concatenation
people, and thus, there is no way to know who the actual ⊕ XOR operation
user is. Therefore, it cannot provide non repudiation. Hence,
biometric keys have been proposed [5], which are based on 2.1 The Registration Phase
personal characteristics such as fingerprints, palm prints,
and irises. Before the users login to the system, they must perform
Li and Hwang have proposed an efficient biometrics- the following steps, as shown in Figure 1.
based remote user authentication scheme using smart cards Step 1: The users offer their personal biometrics, Bi , on
[6], which, however, could not withstand the impersonation the specific device and input the password, PWi , and the
and power consumption attacks. This paper shall point out user identity, IDi , to the registration center in person.
the flaws of this proposed scheme.
The rest of this paper is organized as follows. In Section
2 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

Step 2: The registration center computes the messages Step 1: Si checks if the format of C i ’s IDi is valid or
ri = h ( PWi || fi ) and not.
ei = h( IDi || Xs) ⊕ h( PWi || fi ) , where fi = h( Bi ) and Xs is Step 2: If the above-mentioned holds, Si computes the
the secret information generated by Si . messages M 3 = h( IDi || Xs) , M 4 = M 2 ⊕ M 3 = Rc ,
Step 3: The registration center stores ( IDi , h (.), fi , ei ) M 5 = M 3 ⊕ Rs , and M 6 = h( M 2 || M 4 ) to provide mutual
into the user’s smart card and then sends it to the user authentication between client and server.
through a secure channel. Step 3: Next, Si sends the messages (M5, M6) to Ci .
Step 4: On receiving the Si ’s message, C i checks if
Ci Ri
IDi , Bi , PWi M 6 = h( M 2 || Rc) .
Step 5: If the above-mentioned holds, C i considers that
Computes
Si is authenticated and then computes the following
ri = h(PWi || fi )
messages to offer mutual authentication between client and
ei = h(IDi || Xs) ⊕ h(PWi || fi ) server.
Stores ( ID i , h(.), f i , e i ) in the smart card M 7 = M 5 ⊕ M 1 = Rs ,
M 8 = h( M 5 || M 7 ) ,
Smart card where M7 is the random number of the server. The client,
which knows M 1 = h ( IDi || Xs) , can send back the message
of M 8 = h((h( IDi || Xs ) ⊕ Rs) || Rs) .
Figure 1. The registration phase
Step 6: Ci sends the message M8 to Si .
2.2 The Login Phase Step 7: On receiving C i ’s message, Si checks if
Whenever the users want to login to the server, they need M 8 = h( M 5 || Rs) .
to perform the following steps, as shown in Figure 2.
Step 8: If it holds, the server accepts C i ’s login
Step 1: The users insert their smart card into the smart
card reader of a terminal and offer their personal biometrics, request; otherwise, it rejects it.
Bi , on the specific device to verify user biometrics. Next,
the system checks if h( Bi ) = f i . Ci Si
Checks the format of Ci ' s IDi
Step 2: If it holds, the user passes the biometrics
Computes M 3 = h(IDi | | Xs)
verification. Then the user inputs the PWi . Otherwise, it
M 4 = M 2 ⊕ M 3 = Rc
means the user did not pass the biometrics verification and
the client terminates the session. M5 = M 3 ⊕ Rs
Step 3: After receiving C i ’s password, the smart card M6 = h(M 2 || M 4 )
will compute the messages M5 , M 6
ri ' = h( PWi || f i ) , M 1 = ei ⊕ ri ' = h( IDi || Xs) , and ?
Verifies M 6 = h(M 2 || Rc)
M 2 = M 1 ⊕ Rc , where Rc is a random number generated
Computes M 7 = M 5 ⊕ M1
by the user.
Step 4: Finally, Ci sends the messages ( IDi , M 2 ) to Si . M 8 = h(M 5 || M 7 )

M8
Ci Si
?
Inserts the smart card and offers Bi Verifies M 8 = h(M 5 || Rs)

Verifies h(Bi )=? fi , Ci inputs PWi

Figure 3. The authentication phase
Computes ri' = h(PWi || fi )
M1 = ei ⊕ ri' 2.4 Changing of password
M 2 = M1 ⊕ Rc Whenever the users want to change their passwords, they
can easily and freely change the password PWi to a new
IDi , M 2 password, PWi new , as shown in Figure 4.
Step 1: The users insert their smart card in the smart
Figure 2. The login phase card reader and offer their biometrics to the specific device
in order to verify the user biometrics.
2.3 The Authentication Phase Step 2: If it holds, the user can input the old
On receiving the login request message, Si will password, PWi , and the new password, PWi new .
authenticate whether the user is legal or not in the following Step 3: The smart card computes
manner, as shown in Figure 3. ri′ = h( PWi || f i ) , ei′ = ei ⊕ ri′ = h( IDi || Xs ) , and
 (IJCNS) International Journal of Computer and Network Security, 3
Vol. 2, No. 3, March 2010

ei′′ = ei′ ⊕ h( PWi new || fi ) , the ei will be replaced with ei′′ 3.2 It cannot withstand password-guessing attacks
new
According to some of the research work [7], [8], storing
and PWi has been changed with PWi . of data as smart card messages is vulnerable, because the
secret information stored in the smart card could be
Ci Si extracted by monitoring its power consumption (power
Inserts the smart card and offers Bi analysis attacks). Using power consumption attacks, the
attacker can get the message ( fi , ei ) and intercept the
Verifies h(Bi )=? fi , Ci inputs PWi and PWi new
message (M2, M6) from the network. Through password-
Computes ri' = h(PWi || fi )
guessing attacks, the attacker can guess the password P Wi′ ,
ei' = ei ⊕ ri' = h( IDi || Xs)
and compute the value as Rc' = ei ⊕ h( P Wi′ || fi ) ⊕ M 2 to
ei'' = ei' ⊕ h( PWi new || fi )
check if M 6 = h( M 2 || Rc' ) . If it holds, the attacker can
The ei will be replaced with ei'' masquerade the user. Otherwise, the attacker can try for the
next guessed password P Wi′ until M 6 = h( M 2 || Rc' ) is
true. A detailed description is given below:
Figure 4. Change password phase Step 1: Using power consumption, the adversary gets the
message ( fi , ei ) .
3. The Weaknesses of Li and Hwang’s Step 2: The adversary intercepts the message (M2, M6)
from the network.
Proposed Scheme
Step 3: By choosing a password P Wi′ , the adversary
It can be seen that Li and Hwang’s proposed scheme
computes the message of Rc' = ei ⊕ h( P Wi′ || fi ) ⊕ M 2 to
enables users to change their passwords freely and provides
mutual authentication between the user and the server. The check if M 6 = h( M 2 || Rc' ) .
most significant feature of this scheme is that its operating Step 4: If the above-mentioned holds, it means the
mechanism is based on the users’ personal biometrics. password guessed, P Wi′ , is the correct password; thus, the
However, Li and Hwang’s proposed scheme still retains adversary can masquerade the user.
three weaknesses, as explained below:
Step 5: On the contrary, if it does not hold, the adversary
tries for the next password guess until M 6 = h( M 2 || Rc' )
3.1 It cannot protect against impersonation attacks is true.
In the authentication phase, the server checks the format
3.3 Adversary can impersonate not only the client
of the client’s identity; if it holds, the server computes the
but also the server
message (M3, M4, M5, M6), and the server can get the
message ( IDi , M 2 ) from the client in the login phase. The Through a power analysis attack and the above-mentioned
statement 3.2, the attacker can get the value of h( IDi || Xs) ;
server can use the message ( IDi , M 2 ) and secret
then, the attacker can masquerade not only the client but
information Xs to masquerade M2 ' , then get
also the server. As the attacker can intercept the
M 1 = M 3 = h( IDi || Xs) ; therefore, the attacker can message ( IDi , M 2 ) from the network, using the message
impersonate the client. The detailed procedure is given h( IDi || Xs) , it can compute the message
below:
(M 3 , M 4 , M 5 , M 6 ) to masquerade the server as well.
Step 1: The malicious server can compute the hash value
h( IDi || Xs) itself, and sends the message h( IDi || Xs) to the
adversary.
Step 2: After receiving the message h( IDi || Xs) , the 4. Conclusions
adversary chooses the random number Rc' to compute This paper points out that the scheme proposed by Li and
M 2 ' = h( IDi || Xs) ⊕ Rc ' . The adversary can Hwang is not secure enough against some weaknesses and
masquerade IDi and sends the login request message proves that it is incapable to withstand impersonation and
power consumption attacks. The attacker can break in as a
( IDi , M 2 ' ) to the server. legal user and intercept the messages from networks to
Step 3: The adversary can use the hash value h( IDi || Xs) masquerade the user. Although Li and Hwang’s proposed
to compute the random number Rs chosen by the server. The personal biometrics scheme is practical in some fields, the
adversary will compute M 8 = h((h( IDi || Xs) ⊕ Rs ) || Rs) to security is substandard.
achieve the validity of the login processes. The drawback
exists because the server does not execute the biometrics
Acknowledgment
verification processes, thus causing insider attacks as
mentioned above. Furthermore, the scheme cannot achieve
non repudiation. This work was partially supported by the National
Science Council, Taiwan, R.O.C. under Grant NSC 98-
2221-E-324-019.
4 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

References
[1] L. Lamport “Password authentication with insecure
communication”, Communications of the ACM 24, pp.
770–772, 1981.
[2] M.S. Hwang and L.H. Li, “A new remote user
authentication scheme using smart cards”, IEEE
Transactions on Consumer Electronics, vol. 46, pp.
28–30, 2000.
[3] M. Kim and CK. Koc, “A simple attack on a recently
introduced hash-based strong-password authentication
scheme”, International Journal of Network Security,
vol. 1, pp. 77–80, 2005.
[4] N.Y. Lee and Y.C. Chiu. “Improved remote
authentication scheme with smart card”, Computer
Standards and Interfaces, vol. 27, pp. 177–180, 2005.
[5] C.T. Li and M.S. Hwang, “An online biometrics-based
secret sharing scheme for multiparty cryptosystem
using smart cards”, International Journal of Innovative
Computing Information and Control, 2009.
[6] C.T. Li and M.S. Hwang, “An efficient biometrics-based
remote user authentication scheme using smart cards”,
Journal of Network and Computer Applications, vol. 33,
pp. 1-5, 2010.
[7] T.S. Messerges, E.A. Dabbish, and R.H. Sloan,
“Examining smart-card security under the threat of
power analysis attacks”, IEEE Transactions on
Computers, vol. 51, pp. 541–552, 2002.
[8] P. Kocher, J. Jaffe, and B. Jun, “Differential power
analysis”, Proceedings of Advances in Cryptology, pp.
388–397, 1999
 (IJCNS) International Journal of Computer and Network Security, 5
Vol. 2, No. 3, March 2010

New Illumination Compensation Method for Face

Recognition
Heng Fui Liau1and Dino Isa2
1
Faculty of Engineering, School of Electrical and Electronic Engineering,University of Nottingham Malaysia Campus,
Jalan Broga, 43500 Semenyih, Selangor, Malaysia.
[email protected]
2
Faculty of Engineering, School of Electrical and Electronic Engineering,University of Nottingham Malaysia Campus,
Jalan Broga, 43500 Semenyih, Selangor, Malaysia.
[email protected]

of a lambertian surface in fixed pose, under variable lighting

Abstract: This paper proposes a method for implementing
illumination invariant face recognition based on discrete cosine
and without shadowing, is a 3D linear subspace [5]-[7]. One
transform (DCT). This is done to address the effect of varying of the approaches used to solve the problem is building a 3D
illumination on the performance of appearance based face illumination model. Basri et al [5] proposed spherical
recognition systems. The proposed method aims to correct the harmonic model where illuminated images are represented
illumination variation rather than to simply discard it. In the in low-dimensional subspace. Lee et al [8] proposed the 9D
proposed method, illumination variation, which lies mainly in linear subspace approach using nine images captured under
the low frequency band, is normalized in the DCT nine different lighting directions. Face images obtained
domain. Other effects of illumination variation which manifest under different lighting conditions are regarded as a convex
themselves by the formation of shadows Lambertian and can be approximated well by a 9D linear
and specular defects are corrected by manipulating the subspace [8]. The illumination convex cone method [9]-[11]
properties of the odd and even components of the DCT. The
showed that a set of images of an object in a fixed
proposed method possesses several advantages. First, it does not
pose under all possible illumination conditions is a convex
require the use of training images or an illumination
model and can be applied directly to the test image. Second is cone in the image space. This method requires a number of
the simplicity of the system. It needs only one parameter to be training images for each face taken with different lighting
determined, making it simple to directions to build the generative model. Unfortunately, the
implement. Experimental results on Yale face database B using illumination cone is extremely complex to build. Several
PCA- and support vector machines (SVM)-based face simpler models that approximate it in the best possible way
recognition algorithms showed that the proposed method gave with minimum complexity is presented in [10], such as low
comparable performance compared to other well-known but dimension subspace model, and cones-attached and cones-
more complicated methods. cast using extreme rays. Gross et al [12] proposed a 3D
Keywords: face recognition, illumination invariant, discrete method based on light-field estimation. It operates by
cosine transform, biometrics. estimating a representation of the light field of the subject
head. The light field is then used as the set of features for
recognition. Recently, Zhang et al [13] proposed a 3D
1. Introduction Spherical Harmonic Basis Morphable Model. They showed
that any faces under arbitrary unknown lighting condition
Face recognition has gained much attention in the past two can be simply represented by three low-dimensional vectors
decades, particularly for its potential role in information and which correspond to shape, spherical harmonic and
forensic security. Principal component analysis illumination. One of the major drawbacks of the 3D
(PCA) [1] and linear discriminant analysis (LDA) [2] are illumination model-based approach is that a number
the two most well-known and widely-used appearance-based of training images of the subject under varying lighting
methods. Existing face recognition systems including PCA conditions or information of 3D shapes are needed during
and LDA do not perform well in the presence the training phase. Moreover, its application range is
of illumination variation. For example, Adini et al [3] stated limited by the 3D face model, and has significant drawback
that the variation between the images of the same person for real-time system due to its expensive computation
due to illumination inconsistencies is always larger than requirement.
image variation due to the change in face identity. Due to Every single pixel of the illuminated face image can be
the importance of this issue, recent researchers focus on regarded as a product of reflectance and luminance at that
developing robust face recognition systems. However, point. Luminance varies slowly while reflectance can
illumination variation still remains a challenging problem change abruptly. In other words, luminance, which
[4]. corresponds to the illumination variation, is located
The problems associated with Illumination variation for mainly in low frequency band but reflectance, which
facial image are mainly due to the different appearance of corresponds to the stable facial features under illumination
the 3D shape of human faces under lighting from different variation, is located at higher frequency band. Land's
direction. Early work showed that the variability of images "retinex” model [14] approximates the reflectance as the
6 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

ratio of the image and its low-pass version serves as zero mean and unit variance. The advantage of the method
estimation for luminance. Chen et al [15] discards the low is low computation complexity.
frequency DCT component in the logarithm domain to This paper proposes an new illumination compensation
achieve illumination invariance. They first expand the pixel method for face recognition . The proposed method aims to
intensity in the dark region using logarithm transform. correct the illumination variation rather than to simply
Then, discrete cosine transform is applied on the image in discard it. Compared to existing methods based on the DCT,
logarithm domain. The low frequency components are the proposed method here does not discard the low
discarded by setting the corresponding coefficients as zero. frequency components which represent most of the effects of
Nanni and Lumini [16] had investigated discrete wavelet illumination variations. This allows us to retain more
transform features under different lighting conditions. They information, which might be useful for face
found that the most useful sub-band are the first level of recognition. Furthermore, some of the effects of this
decomposition obtained by Biorthogonal wavelet and the illumination variation lie in the higher frequency band,
horizontal details are the third level of decomposition approximately the same frequencies as occupied by some
obtained by Reverse Biorthogonal wavelet, which is robust important facial features. As previously mentioned, the
against illumination variation. Franco and Nanni [17] effect of these kinds of illumination variation creates
proposed classifiers fusion scheme. PCA is applied to shadow and specularities in the image. Thus, removing the
extract and reduce the dimensionality of the useful features low frequency component does not help in this case. The
from DWT domain [16] and DCT domain [15]. The proposed method here manipulates the odd and even DCT
individual classifier for each feature is Kernel Laplacian components to remove these artifacts. Using the DCT
Eigenmap. The output of each classifier is further fused approach, the complexity of the previously described
using sum rule. methods is avoided while retaining a comparable
Generally, human face is similar in shape, have two eyes, performance [8]-[10], [15], [21]-[24]. The proposed method
a mouth and a nose. Each of these components makes is based solely on 2D images and does not need to estimate
different distinctive shadows and specularities depending on the 3D shape. It requires much less computation resource
the direction of the lighting in a fixed pose. By using such than other methods based on 3D model as stated above.
characteristic, the lighting direction can be estimated and Besides that, it does not require any prior information on the
illumination variation can be compensated. Quotient image illumination model and training images. Furthermore, the
(QI) [18] is an effective method to handle illumination parameter selection is simple. Only one parameter, the cut-
problem. This method is very practical because it off-frequency of the filter, is required to be
only requires one training image for each person. The determined. Experimental results on Yale face database B
authors in literature [19] and [20] further improved the [11] using PCA- and support vector machines (SVM)--based
original QI method and proposed the self-quotient face recognition algorithms showed that the proposed
image (SQI) method. The luminance is estimated as the method has achieved good performance as compared
smoothed version of the image by Gaussian filtering. The to some of the more complicated methods described above.
illumination variation is eliminated
by dividing the estimated luminance. However, the 2. Method
parameter selection for the weighted Gaussian filter is
empirical and complicated. Shan et al As aforementioned before, an illuminated face image, I(x,y)
[21] proposed quotient illumination relighting (QIR) , can be regarded as product of reflectance, R(x,y) , and
method which synthesizes images under a predefined luminance, L(x,y) , as shown in (1). Taking logarithm
normal lighting condition from the provided face images transform on (1), we have (2).
captured under non-uniform lighting conditions. Zhao et al
[22] proposed illumination ratio image. One training image
for each lighting conditions is required to simulate the
distribution of the images under varying lighting condition.
Liu et al [23] proposed an illumination restoration method. (2)
The illuminated image is restored to image under frontal
light source using a ratio-image between the face image The logarithm transform transforms (1) into linear
under different lighting conditions and a face image under equation where the logarithm transform of the illuminated
frontal light source, both of which are blurred by Gaussian image is the sum of the logarithm transformed of reflectance
filter. The image is further enhanced through an iterative and the logarithm transform of luminance as shown in
algorithm where Gaussian filter with different sizes are (2). In the image processing field, logarithm transform
applied on different regions of the face image. Vuccini et al is often employed to expand the values of the dark
[24] adopted Liu et al’s method to restore and enhance the pixels. Fig. 1 gives a comparison between the original image
illuminated image. LDA is employed as feature extractor. and the same image following the logarithm transform. The
They proposed an image synthesis method based on QI to brightness of the right image is spread more uniformly.
generate training images to overcome the small-sample-size
problem. Xie and Lam [25] normalize the lighting effect
using local normalization technique. The face image is
partitioned to blocks with fixed size. For each block, the
pixel intensities are normalized in such a way that it has
 (IJCNS) International Journal of Computer and Network Security, 7
Vol. 2, No. 3, March 2010

number. Furthermore, DCT has lower computation

complexity than Discrete Wavelet Transform. For an N×N
image, the 2D DCT is given by (5). For υ , ν = 0,1,2,…N-1.
and represent the horizontal component and vertical
component respectively is the is the value of the
pixel at coordinate (x,y). For 2D DCT, the frequency is
computed as the square root of the sum of and . The
inverse 2D DCT is expressed as (6).

Figure 1. On the left is the original face image. On the right

is the output of logarithm transform (5)

Luminance component lies mainly in the low frequency

band while the reflectance component lies mainly in the
higher frequency band. The luminance component can be
approximated as the 'low-pass version' of the illuminated From (6), the uniform luminance component can be
image. The reflectance component of an illuminated face obtained from the original image by adding a compensation
image represents the stable facial features under varying term, which compensates for the nonuniform
lighting conditions. Thus, the illumination variations can be illumination.
compensated for provided we can estimate them relatively Generally human faces are almost symmetrical. The left-
well. half side of the face is almost identical to the right-half side
Let be the incident luminance and be the of the face in most of the cases. For non-uniformly
uniform luminance. A uniformly illuminated image, , can illuminated images, part of the face is brightened and the
be expressed as (3). In order to normalize the brightness others are darkened. The component is
across all of the images, the mean value of are set to approximated as the 'low-pass version' of with
be the same for all image. certain cut-off frequency. The compensation term can be
estimated as follow. Firstly, the mean of an image
(3) reconstructed from low frequency DCT components is
computed according to (7). Then, each pixel is subtracted by
the mean as shown in (8). is estimated using the
low frequency DCT components. A negative value indicates
The luminance component corresponds to the illumination that a single pixel is 'dark' and positive value indicates that
variation. As previously described, the illumination the corresponding pixel is 'bright'. Each pixel is adjusted by
variation can be well-approximated by the low frequency halving the difference between pixel value and mean value
components of the illuminated face image. In section 2.1, an as shown in (8). In other words, the intensity of each pixel is
illumination compensation method based on low frequency adjusted according to the difference between the mean
DCT components are presented. In fact, illumination intensity of the image that is reconstructed using only low
variation cannot be perfectly separated from the original frequency components. The illumination variations which
image in the frequency domain. Under large and complex lie mainly in the low frequency band are reduced while
illumination variations, the shadows and specularities are important facial features are preserved. In other words,
casted on the face due to the unique 3D shape of human the low-pass version of the image in the DCT domain serves
face. Some shadows and specularities lie in the same as an estimated of the luminance component. Fig. 2 shows
frequency band as the reflectance, which corresponds to the effect of illumination normalization in the low frequency
facial features do. In this case, compensation technique DCT components on a face image. The left-most image is
based on low frequency components will be unable to the original image. The second, third and forth images are
remove the artifacts. In section 2.2, an illumination the processed image with the cut-off frequency set at 2, 4
compensation method based on the properties of odd and and 6 respectively. As you can see, the shadow effects at left
even DCT components in horizontal direction, which has side of the face are reduced at the increasing value of the
the capability to remove artifacts in the high frequency cut-off frequency while the facial features are unaffected.
band, is presented. Unlike other similar methods, only one parameter needs to
. be determined, which is the cut-off frequency of the low-
pass filter. However, the normalization may cause some
2.1 Illumination normalization using low frequency
information lost in the process due to wrong estimations of
DCT components
the 'dark' and 'bright' zones.

DCT has been widely used in modern imaging and video

compression. DCT possesses some fine properties, such as
de-correlation and energy compaction. The DCT is chosen
because it only gives real value, unlike Discrete Fourier
Transform (DFT) which gives complex
8 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

is always below 0.5. For the non-uniformly illuminated

image, the strong signals (greater than one) are located at
the low frequency band. However, the middle and high
frequency band contain points that have strength greater
than 0.5. This shows that the illumination variation also
corrupts the high frequency component, which corresponds
to reflectance. Hence, this proves that no features are robust
enough to be against the illumination variation.
Figure 2. Normalization on low frequency DCT
components. The left most image is the original image

2.2 Illumination Correction Using Odd and Even DCT

Components in Horizontal Direction
Under large and complex illumination variation, the effect
of illumination variation occurs across almost the
entire frequency band. Some variations such as shadows Figure 4. The top and bottom is the odd DCT component of
and specularities lie in the same frequency band as some original image and illuminated image respectively
important facial features do. In this section, we
propose an illumination correction method which uses the
Since human face generally have symmetry property, we
properties of odd and even DCT components to eliminate
can easily compensate the illumination variation in the face
the shadows and specularities and to restore the original
based on the properties of odd and even components of
features. DCT. Here, an illumination compensation method based on
manipulating even and odd DCT components is proposed.
First, two new images, and , are reconstructed
from the odd and even DCT components in the horizontal
direction of the original images. The pixels on the left half
Figure 3. 2D-DCT basis functions are compared with the pixels on the right half. If the left-
side pixel is negative and the corresponding right-side pixel
is positive, the values of both pixels will be adjusted
In Fig. 3, different images have been produced by varying according to (10) and (11). is the compensation term. Fig
the DCT components in the horizontal direction. As shown .5 shows an illuminated image and the corresponding
in the Fig. 3, the second and forth images have and .
been reconstructed by different even DCT components in the
horizontal direction and the symmetry is evident. The first
and third asymmetry images are reconstructed by odd DCT
components in the horizontal direction
are asymmetrical. Non-uniform illumination gives rise
to the odd DCT components and affects the
magnitude of the even DCT components. (5) can now be
rewritten in that,
Figure 5. From left to right: illuminated image, and

Odd DCT components can be regarded as noise or

luminance components due to illumination variation and the
even DCT components represent the wanted information
which may or may not be slightly corrupted by non-uniform
luminance. Under such model, the luminance components
can be estimated using odd DCT components thereby giving
estimation for the correction term as well. Fig. 4 shows
the difference between odd DCT components of image under If the left-side pixel is positive and the corresponding
variable illumination and the image of same person under right-side pixel is negative, the values of both pixels will be
uniformly illuminated condition. As shown in the top of adjusted according to (12) and (13).
Fig. 4, the odd components of the illuminated images
fluctuate greatly with the magnitude greater than one. In
contrast, the bottom of Fig. 4 also shows that illumination
variation happens across the entire frequency band. Note
that, the signal strength of the uniformly illuminated image
 (IJCNS) International Journal of Computer and Network Security, 9
Vol. 2, No. 3, March 2010

3. Experiment results on Yale Face database B

Yale face database B is commonly used to evaluate the
performance of illumination invariant face recognition. The
Yale face database B contains 10 individuals in nine
different poses. For each pose, there are 64 different
illumination conditions. All images were manually aligned.
The size of face images are cropped to 64×55 pixels. The
face images were divided into 4 subsets according to the
[11].The four subsets were, subset 1 ), subset 2
Figure 6. The sum of the correction term and even DCT
), subset 3 ) and subset 4 ),
components gives the illumination corrected image
where bracketed numbers indicate the angular position
of light of incident with respect to the person.
The new pixel intensity for the compensated image can
then be computed using (14). As shown in Fig. 6, the
3.1 Face recognition using principal component
shadow created by the nose has been eliminated. Fig. 7
analysis
shows the illumination-corrected image using the proposed
method. Illumination variations are largely reduced while
the important facial features are preserved. Here, we would PCA is a classical face recognition method. PCA seeks for a
like to emphasize that the proposed method is able to set of projection vectors which projects the input data in
recover facial feature as long as the specularities or shadow such a way that the covariance matrix of the training images
regions occur only on one side of the face. The output image is maximized. Six images in subset 1 were selected to serve
looks different from the original image because the proposed as training images to build the covariance matrix for PCA
method corrects the illumination variation based on the [13] and [17]. Eigenvalue decomposition was applied on the
symmetry of human face. The output image is aligned in covariance matrix to seek for the eigenvector that projected
such a way that the left-half side is identical to the right-half the data in the direction that maximized the covariance
side. Due to some subtle variations in the two halves of matrix. The nearest neighbour rule was used as a classifier
every face, a face image is not exactly symmetrical; for with the Euclidean distance as the distance metric. Fig. 8
example, the two eyes have slightly different sizes, there is a shows that the largest thirty principal components were
mole at only one side of cheek, etc. Therefore, the output sufficient to give the best result and to represent the face
images may have an adverse effect. All of the images are image. Therefore, thirty principal components were used to
normalized to a mean of 0.5. form the projection matrix. Fig. 9 and Fig. 10 show the
performance of the proposed method for subset 3 and 4
respectively under different values of cut-off frequency.
Setting the cut-off frequency at six gave the best results for
both subsets.

Figure 7. Illumination normalization and compensation

Figure 8. Recognition rate for subset 3 and subset 4 that
based on proposed method
correspond to the number of principal components

The flow of the proposed illumination invariant face

recognition system can be summarized as below:
1. Perform logarithm transform on the input images to
expand the values of the dark pixels.
2. Correct the non-uniform luminance in the low frequency
domain using the proposed method in section 2.1.
3. Eliminate the specularities and shadows that lie in the
same frequency band as the reflectance does using the
proposed method in section 2.2.

Figure 9. Recognition rate for subset 3 that correspond to

the cut-off frequency
10 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

of SVM tremendously in Subset 4. Our method scored 100%

recognition rate in Subset 3 and 96.4% in subset 4.

Table 2: Face recognition using SVM

Method Subset 3 Subset 4
Original 13.6% 55%
Histogram 100% 61.4%
equalization
Proposed method 100% 96.4%
Figure 10. Recognition rate for subset 4 that correspond to
the cut-off frequency
3.3 Comparison with other methods
The results for subset 3 and 4, along with other well-
Table 1: Face recognition using PCA
known methods, are presented in Table 3. From Table 3,
Method Subset 3 Subset 4 it is shown that the proposed
Histogram 58.4% 15.6% method gave results comparable with other well-known but
equalization more complicated methods. In contrast, the proposed
Proposed method 100% 92.1% method is much simpler to implement with much less
computation resource requirement. The baseline system only
Without illumination normalization in the low frequency scored 58.4% and 15.6% in subset 3 and subset 4
band, the proposed method was only able to achieve 96.3% respectively. The proposed method outperformed linear
and 64.1% recognition rate in subset 3 and 4 respectively. subspace method, cones-attached method, illumination ratio
The poor results were due to the large variation in the low image method and QIR method. As aforementioned, the
frequency band where the illumination correction algorithm above stated methods required training images to build the
was unable to cope with. As we increased the value of cut- illumination model or to estimate the lighting
off frequency, the performance improved steadily although conditions. The authors in [15] discards low frequency
there were some small fluctuations. It gave the best components, keeping only the high frequency ones
performance when the cut-off frequency was set at 6 (100% which correspond to the reflectance component. Fig.
for subset 3). For subset 4, the best performance was 92.1% 11 shows the difference between the proposed method and
when the cut-off frequency was at 6, 7 or 9. Beyond that, the [15]. The shadows and specular defects remained on the
performance decreased because some important image because they lay in the same frequency band as the
features might have lost in the illumination normalization reflectance component does. Discarding low frequency
process. components might remove the important facial features that
Table 1 shows that our method can achieve a better lie in low frequency band and degrade the performance of
performance level than if only histogram equalization was the face recognition system when the number of class was
used. large. The proposed method aims to correct the
illumination variation rather than to simply discard it to
3.2 Face recognition using support vector machine achieve illumination invariance. However asymmetrically,
it was unable to completely remove artifacts that appeared
Support vector machine (SVM) [26]-[28] is a popular
on the both sides of the face. This caused the proposed
technique for classification motivated by the results of
method to have rather unimpressive performance in subset
statistics learning theory. Unlike traditional methods such as
4 compared to [15]. Our method is robust against
neural network that minimize the empirical training error,
illumination variation and is easy to implement. It can be a
SVM is designed based structural minimization principal.
pre-processing stage for other face recognition algorithms.
SVM maps the training data into a higher dimensional
feature using kernel trick, in which an optimal hyperplane
Table 3: Recognition rate comparison with other methods
with large separating margin between two classes of the
Methods Recognition rate
labeled data is constructed. The training data of SVM are
(%)
the PCA feature set described in the previous section. In this
paper, several types of kernel such as linear kernel, wavelet subset 3 subset 4
kernel, polynomial kernel and radial basis function (RBF) Linear Subspace [10] 100 85
kernel are studied. RBF kernel gives the best result. The Cones-attached [10] 100 91.4
RBF function is defined as below: Cone-cast [10] 100 100
Illumination ratio image [22] 96.7 81.4
Quotient Illumination Relighting 100 90.6
[21]
illumination restoration [23] 98.3 96.4
where is the width of the RBF. DCT in logarithm domain [15] 100 99.8
Vucini et al’s method [24] 100 95
Table 2 shows the performance of SVM under different Our method+PCA 100 92.1
types of methods. Our algorithm improved the performance Our method +SVM 100 96.4
 (IJCNS) International Journal of Computer and Network Security, 11
Vol. 2, No. 3, March 2010

[5] R. Basri, D.W. Jacobs, “Lambertian reflectance and

linear subspaces”, IEEE Transaction on Pattern
Analysis and Machine Intelligent, 25(2), pp.218-233,
2003.
[6] S. Nayar, H. Murase, “Dimensionality of illumination
manifold in eigenspace”, Technical Report CUCS-021-
94 Columbia University.
[7] S. Amnon, “Photometric issues in 3D visual
recognition from a single 2D image”, International
Figure 11. Comparison between proposed method (bottom)
Journal of Computer Vision,2, pp. 99-122, 1997.
and [10] (top)
[8] K.C. Lee, J. Ho, D.J. Kriegman, “Nine points of light:
acquiring subspaces for face recognition under variable
3.4 Computation time
lighting and pose”, In Proceedings of the IEEE
An efficient and simple illumination compensation method Conference on Computer Vision and Pattern
in DCT domain is presented. All experiments were Recognition, vol.2, pp. 519-526, 2001.
conducted using Matlab 2008a on Core 2 Duo E6750 CPU [9] P.N. Belhumeur, D.J. Kriegman, “What is the set of
and 2GB RAM. It takes 2.050 seconds and 2.311 seconds to images of an object under all possible illumination
process all of the face images included in subset 3 and conditions?”, In Proceedings of the IEEE Conference
subset 4 respectively. The mean computation time for each on Computer Vision and Pattern Recognition, pp. 270-
image is 0.0167 second. The complexity is low and it can be 277, 1996.
implemented in real-time system. [10] A.S. Georghiades, P.N. Belhumeur, D.W. Jacobs,
“From few to many: illumination cone models for face
4. Conclusion recognition under variable lighting and pose”, IEEE
Transaction on Pattern Analysis and Machine
This paper proposes a novel illumination compensation
Intelligent, 23(6), pp.630-660, 2001.
method in DCT domain. The illumination is normalized
[11] H.F. Chen, P.N. Belhumeur, D.J. Kriegman, “In search
based on the low frequency components of DCT.
of illumination invariants”, iIn Proceedings of the
Subsequently, the illumination variations which create
IEEE Conference on Computer Vision and Pattern
shadows and specularities are further corrected by the
Recognition, pp. 13-15, 2000.
proposed method which uses the properties of odd and even
[12] R. Gross, I. Matthews, S. Baker, “Appearance-based
components of DCT. The proposed method is simple in
face recognition and light-fields”, IEEE Transaction
terms of design, where there is only one parameter that
on Pattern Analysis and Machine Intelligent,26(4),
needs to be determined, which is the cutoff frequency of the
pp.449-465, 2004.
filter for the illumination normalization process. The
[13] L. Zhang, S. Wang, D. Samaras, “Face synthesis and
proposed method gives comparable result with other well-
recognition from a single image under arbitrary
known but more complicated methods in Yale face database
unknown lighting using a spherical harmonic basic
B. The proposed method has rather unimpressive
morphable model”, In Proceedings of the IEEE
performance in subset 4 due to reason that large
Computer Society Conference on Computer Vision and
illumination variation creates shadows and specularities that
Pattern Recognition, vol.2, pp.209-216, 2005.
occur at both sides of the face. Our method is robust against
[14] E.H. Land, J.J. McCann, “Lightness and retinex
illumination variation and is easy to implement. It can be a
theory”, Journal of the Optical. Society of America,61
pre-processing stage for other face recognition algorithms.
(1), pp. 1-11, 1971.
[15] W. Chen, M.J. Er, S. Wu, “Illumination compensation
References and normalization for robust face recognition using
discrete cosine transform in logarithm domain”, IEEE
[1] M.A. Turk, A.P. Pentland, “Eigenface for face Transaction on System, Man and Cybernetics. B,36 (2)
recognition”, Journal of Cognitive Neurosience, 3, pp. pp. 458-466, 2006.
71-86, 1991. [16] L. Nanni, A. Lumini, “Wavelet decomposition tree
[2] P.N. Belhumeur, J.P. Hespanha, D.J. Kriegman, selection for palm and face authentication”, Pattern
“Eigenface versus Fisherface”, IEEE Transaction on Recognition Letters, 329, pp.343-353, 2008.
Pattern Analysis and Machine Intelligent,19 (6), [17] A. Franco, L. Nanni, “Fusion of classifiers for
pp.711-720, 1997. illumination robust face recognition”, Expert Systems
[3] Y. Aldini, Y. Moses, S. Ullman, “Face recognition: the with Applications: An International Journal, 36,
problem of compensating for changes in illumination pp.8964-8954, 2009.
direction”, IEEE Transaction on Pattern Analysis and [18] S. Amnon, R.-R Tammy, “The quotient image: class-
Machine Intelligent, 19 (7), pp.721-732, 1997. based re-rendering and recognition with with varying
[4] P.J. Phillips, W.T. Scruggs, A.J. O’Toole, P.J. Flynn, illuminations”, IEEE Transaction on Pattern Analysis
K.W. Bowyer, C.L. Schott, M. Sharpe, “FRVT 2006 and Machine Intelligent,23 (2),pp.129-139, 2001.
and ICE 2006 large-scale results”, National Institute of [19] H. Wang, S.Z. Li, Y. Wang, “Generalized quotient
Standards and Technology Report. image”, In Proceedings of the IEEE Conference on
12 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

Computer Vision and Pattern Recognition, 2, pp. 498- [24] E. Vucini, M. Gökmen, E. Gröller, “Face recognition
505, 2004. under varying illumination”, In Proceedings of the 15th
[20] H. Wang, S.Z. Li, Y. Wang, “Face recognition under Conferences in Central Europe on Computer Graphics,
varying lighting conditions using self quotient image”, Visualization and Computer Vision, pp.57-64, 2007.
In Proceedings of the IEEE International Conference [25] X. Xie, KM. Lam, “An efficient illumination
on Automatic Face and Gesture Recognition, pp. 819- normalization method for face recognition”, Pattern
824, 2004. Recognition Letters. 27, pp.609-617, 2006.
[21] S. Shan, W. Gao, B. Gao, D. Zhao, “Illumination [26] V. Vapnik, The nature of statistical learning theory.
normalization for robust face recognition against Springer-Verlag, New York, 1995.
varying lighting conditions”, In Proceedings of the [27] B.E. Boser, I. Guyon, V. Vapnik, “A training
IEEE Workshop on AMFG, pp. 157-164, 2003. algorithm for optimal margin classiffers”, In
[22] J. Zhao, Y. Su, D. Wang S. Luo, “Illumination ratio Proceedings of the Fifth Annual Workshop on
image: synthesizing and recognition with varying Computational Learning Theory, pp144-152, 1992.
illuminations”, Pattern Recognition Letters. 24, [28] C. Cortes, V. Vapnik, “Support-vector network.,
pp.2703-2710, 2003. Machine Learning”,20 , pp.273-297, 1995.
[23] DH. Liu, KM Lam, LS. Shen, “Illumination invariant
face recognition”, Pattern Recognition. 38 , pp.1705-
1716, 2005.
 (IJCNS) International Journal of Computer and Network Security, 13
Vol. 2, No. 3, March 2010

Towards In-Car Ad Hoc Network for Pervasive

Multimedia Services
Kamal Sharma1, Hemant Sharma2 and Dr. A K Ramani3
1
Institute of Computer Science & IT, Devi Ahilya University,
Indore (M.P.), India
[email protected]
2
Software Architect, Delphi Delco Electronics & safety Europe GmbH,
Tec Centre, D 31162 Bad Salzdetfurth, Germany
[email protected]
3
Prof. & Head, 1Institute of Computer Science & IT, Devi Ahilya University,
Indore (M.P.), India
[email protected]

Abstract: Pervasive computing systems inside modern However, the realization of a fully wireless car bus system is
automobiles are made up of hundreds of interconnected, often still far away. Fiber optic connections for multimedia bus
replaceable devices and software components. In-vehicle systems offer advantages regarding costs and bandwidth,
multimedia components and applications are becoming complex and the demanded reliability of mission-critical networks
artifacts due to advancement in technology and increased still require wired connections to ensure the safe operation
competition. There is a growing need for software platform to
enable efficient deployment of multimedia services in
of the car. Though, cost-effective wireless subsystems which
automotive environment. This paper presents the architecture of could extend or partly replace wired bus systems are already
a Bluetooth based multimedia ad hoc network in a car. The nowadays conceivable. A very promising technology in this
architecture is explained and presented together with a prototype context is specified by the recent Bluetooth 3.0 [1] standard.
implementation running on hand-held devices. Further, the
different fundamental architectural tradeoffs, based on The provisioning of multimedia streaming applications
measurements, have also been analyzed. using wireless network, inside the vehicle, requires
managing differentiated performance levels depending on
Keywords: Ad Hoc Networks, Pervasive Computing, In application/user/device requirements in order to properly
Vehicle Multimedia, Bluetooth. allocate network bandwidth, especially the limited one
available in the wireless last-meter [3]. In particular, the
1. Introduction Bluetooth specification [1] offers limited support to
performance differentiation, by allowing to choose which of
During the last few years, the proliferation of miniaturized the three kind of logical transports to exploit and to
devices with networking capabilities has provided the statically configure performance requirements for ACL
technological grounds for pervasive networking even in ones.
automotive environments. Growing demand for personal
In addition, current implementations of the Bluetooth
device connectivity, mobile Internet access, remote
software stack do not allow applications to exploit the
monitoring and diagnostics, as well as enhanced safety and
limited performance functions included in the specification
security is driving vehicle manufacturers and suppliers to
in a portable way. The result is that the development of
seek out new wireless technologies. Wireless technology
Bluetooth operations in multimedia ad hoc applications
integration strategies would enhance the value proposition
currently depends on specific implementation details of the
of vehicles by integrating advanced electronics systems such
target Bluetooth hardware/software platform. This
as infotainment systems, safety and stability systems, and
relevantly complicates service design and implementation,
comfort and convenience enhancement systems.
limits the portability of developed applications, and calls for
The evolutionary development of in-car electronic systems the adequate modeling of performance parameters
has lead to a significant increase of the number of corresponding to potential ad hoc applications and services.
connecting cables within a car. To reduce the amount of
The design presented in this paper is for a wireless
cabling and to simplify the interworking of dedicated
streaming system that offers a means for bringing the
devices, currently appropriate wired bus systems are being
participatory media and bulk content distribution into the
considered. These systems are related with high costs and
wireless domain. The basis for the service is an
effort regarding the installation of cables and accessory
opportunistic distribution of among network. Users in our
components. Thus, wireless systems are a flexible and very
system exchange data when their corresponding
advanced alterative to wired connections.
application(s) receive trigger. The system is going to be
14 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

open to any node who wants to provide and consume the

content. Hence, it is based on unlicensed short-range
communication specifically Bluetooth.

Figure 1. Overview of In- Car Multimedia Network

By relying on short-range communication, the network will 2.1 The Multimedia PAN
be highly disrupted most of time. The communication is
further challenged by relatively short transfer opportunities Bluetooth technology is based on a master-slave concept
which might be in the range of a few seconds when for where the master device controls data transmissions through
example two nodes communicate with each other. The a polling procedure. The master is defined as the device that
contribution of this paper is twofold: initiates the connection. A collection of slave devices
associated with a single master device is referred to as a
Ø A mechanism for content streaming based on piconet. The master dictates packet transmissions within a
opportunistic communication; piconet according to a time-slot process. The channel is
divided into time slots that are numbered according to an
Ø An evaluation of the mechanism through the internal clock running on the master. A time division
use of realistic use cases and real traces.
duplex (TDD) scheme is used where the master and slaves
The rest of the paper is organized as follows: Section 2 alternatively transmit packets, where even numbered time
provides the description of system and potential application slots are reserved for master-slave transmissions, while odd
scenarios. In section 3, the architecture and design of the numbered time slots are reserved for slave-master
system has been discussed. Section 4 evaluates the design of transmissions.
the system. Section 5 provides an overview of related
research and section 6 concludes the paper.

2. System Description

An ad hoc multimedia network in a car and participating

devices are shown in figure 1. The diagram presents an ad
hoc network containing a smart phone, an infotainment
system, and rear seat entertainment system. The devices are
equipped with Bluetooth module and are capable of
establishing a Bluetooth connection. A piconet can be
established between Infotainment system and iPhone when
the driver or any other occupant of the vehicle pairs the
phone. Similarly a piconet could be established between
Infotainment system and rear seat entertainment system.
The piconet shall enable sharing of iPhone data or access to
information from internet, if appropriate application
framework is available at infotainment system.
Figure 2. The Multimedia PAN
 (IJCNS) International Journal of Computer and Network Security, 15
Vol. 2, No. 3, March 2010

Figure 2 provides an overview of the multimedia PAN up and running, its Bluetooth module is ready to pair with
inside the car with the role of different nodes. The piconet other available device. Pairing with iPhone or with rear-seat
here consists of a master, the infotainment device, and two entertainment unit or with both establishes the network.
slaves, the smart phone (iPhone) and rear-seat
entertainment device. As soon as the infotainment system is

Figure 3. Overview of Ad Hoc Multimedia Communication Architecture.

communication of intended data streams. Therefore, every

2.2 Application Scenarios
application subscribes to the channels that it is interested in,
The ad hoc multimedia network consisting of Infotainment and his corresponding device node will try to retrieve any
system and the slave devices shall provide following data belonging to those channels. Following the approach
services: used in the Internet-based podcasting protocols; the
framework structures channels into different media streams.
Ø Access to audio streams from iPhone to To make efficient use of contacts with a small duration, the
infotainment system that could be played and streams are further divide into data packets, transport-level
hearable on the vehicle’s sound system. data units of a size that can typically be downloaded in an
individual node encounter. Each data stream is further
Ø Access to video streams from iPhone to rear- seat divided into protocol data units (PDU), the atomic transport
entertainment unit via the piconet master. unit of the network.
Ø Access to internet from rear-seat unit using iPhone
via the piconet master. The resulting system architecture is illustrated in Figure 3.
The transport layer acts directly on top of the link-layer
Ø Applications based on information received by the without any routing layer. To distribute information among
iPhone to help safe driving, such as weather the communicating nodes, the framework does not rely on
information or traffic information. any explicit multi-hop routing scheme. Instead of explicitly
routing the data through specific nodes, it relies on a
receiver-driven application-level dissemination model,
3. System Architecture and Implementation
where data content are routed implicitly as nodes retrieve
information that they request from neighboring nodes. It
This section describes the design and implementation of our
distinguishes between an application layer, a transport layer,
Ad Hoc network system. We first give a brief overview, then
and the data link layer. In a first level of aggregation, the
describe the system architecture, and finally discuss the
application organizes the data in media stream channels.
protocols involved in a pair-wise association when devices
in the network communicate.
Below the application layer, the transport layer organizes
the data into data streams. Streams are smaller data units
3.1 System Architecture
that should be able to communicate over short contacts. The
The architecture organizes application data into use of smaller file blocks is also supported by the idea of
communication channels to facilitate identification and the integrating forward error correction, for instance the use of
16 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

fountain codes, to speed up and secure the data transfer, synchronization. The transfer mode is based on a client-
specifically when packets are received unordered. The server model. The communication protocol is a request-
streams packets themselves are then again cut into smaller response system. It is not a strict system though; some
parts to optimize the interaction with the data link layer; requests do not generate a response whereas other requests
i.e., the size is set to the PDU size of the data link layer. can generate several responses. The goal is to send as few
data packets as possible to reduce the communication
The proposed system is designed to work on any MAC overhead.
architecture, however, to be effective even in the presence of
short contact durations, short setup times and high data Figure 4 illustrates the state diagram of the transfer mode.
rates are important for achieving high application It shows the three most important stages of synchronization
communication throughput. The design is further between two device nodes: the negotiation, data query, and
characterized by two fundamental choices. data communication stage.

Ø First, it allows only pair wise associations even In the Negotiation stage, both devices determine if some of
when the MAC layer supports multi-point the subscribed channels are available on the other device.
communication. Instead of querying every single channel, the devices
Ø Second, it never pushes data in the network and exchange a channel filter that contains all channel
relies instead on receiver-driven dissemination. subscriptions a device offers. The devices then start to test
their subscribed channels against the filter of the other
The arguments for these two choices are simplicity and
device and create a list of matching channels. This is a local
optimal usage of short contact durations. Furthermore, the
transport layer is able to optimize the flow control between process and does not involve the exchange of messages.
the nodes and is not constrained by the slowest receiver in
range. Since the framework does not perform multi-hop
routing explicitly, the system performance is mainly
determined by the selection of nodes, it is synchronizing
with and the order by which it transfers data from the peers.
This task is performed by the synchronization service,
depicted in Figure 1. The synchronization service is
responsible for maintaining state about past synchronization
encounters and current devices in the PAN.

3.2 Association Phases

This section next describes the process when two nodes

associate. The framework differentiates two phases: a
discovery phase in which nodes detect that they are in a
range and a data exchange phase in which the nodes
negotiate and perform data transfer.

1) Device Discovery Phase

We assume that every device that participates in the

opportunistic data sharing belongs to the same network,
e.g., with IEEE 802.15 every device is configured in ad hoc
mode. While the detection of new devices in the network is
handled by the MAC layer, the application in the nodes has
to take care of the discovery of devices that participate in the
wireless communication service.
Figure 4. State Diagram – Data Transfer Modes.
The synchronization service keeps track of the discovery
messages received from each peer and maintains a history In the Data Query stage, a device confirms the channels
list. An important aspect of the discovery process is to selected in the previous stage and then retrieves a list of
identify peers with a good connection. Since the system is streams offered by the remote device within those channels.
designed for ad hoc scenarios, it has to consider the specifics Our implementation supports three different types of stream
of wireless communication. retrievals:
2) Data Transfer Phase
1. The peer requests any random stream within a
Once a device node has been detected, the synchronization channel that the remote peer offers.
services switches to the transfer mode which handles the
 (IJCNS) International Journal of Computer and Network Security, 17
Vol. 2, No. 3, March 2010

2. A peer requests any stream which is newer than a 4.3 Device Synchronization
given date starting with the newest streams.
We next look at the synchronization time. The
3. The peer requests any streams that are newer than
synchronization time is the time between the moment two
a given date starting with the oldest stream.
devices have discovered each other and associate until they
The actual communication of streams is handled in the Data start data communication. Figure 5 shows the performance
Communication stage which usually takes up most of the of the architecture for device synchronization with different
connection time. The device starts to process the list of PDU number of channels using channel filters compared to the
streams that was created in the previous stage. The time it would take without filters.
communication itself works analogous to the download
process of BitTorrent [7]. Missing PDUs which are available
at the remote peer are randomly selected and downloaded.
Remark that a stream is divided into several pieces (PDUs)
which are requested

4. Evaluation

This section evaluates important design choices of the

proposed architecture as well as the performance we might
expect from a wireless opportunistic application and content
sharing service. Our evaluation relies on our prototype
implementation and focuses on typical deployment scenarios
in which: Figure 5. Device Synchronization with and with out
channel filter.
Ø the nodes are co-located (e.g., all devices are in
the car) ; or 4.4 Data Transfer
Ø When nodes are not co-located (paired iPhone
The system limits the association time to ensure that the
is not in the car).
system remains fair when multiple nodes are interested in
the data provided by just a few nodes. Without this artificial
4.1 Evaluation Setup and Methodology
limit, the first node that associates with the node having
All devices communicated over their integrated Bluetooth new data content would be able to communicate for an
interface (802.15) turned into ad hoc mode. Our data indefinite amount of time, letting the other nodes starve for
data.
communication application is written in C++. The devices
that participate in the data communication system are
configured in ad hoc mode using the same identifier. .

We evaluated the framework’s accuracy by comparing its

data flows to actual Bluetooth flows between the same
devices. Performing this evaluation on a large-scale is
difficult because it requires control over the software
running at both end-points of a Bluetooth flow. We
performed our evaluation by pre-instrumented Bluetooth
stack of two classes of devices: Infotainment Unit and
iPhone.

4.2 Device Discovery Phase

An important aspect of opportunistic wireless podcasting is

the discovery time. The discovery time is the time between
the moment two nodes move into transmission range until
they discover each other at the application layer and start
the synchronization phase.
Figure 6. Average Data Exchange Time between the PAN
This time is directly impacted by the interval at which the devices
messages are sent. The more frequent these messages are
sent, the quicker they will discover each other and be able to To show how our implementation performs on different
transfer data streams. PAN nodes, Figure 6 shows a comparison of the download
18 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

time with the iPhone and the Infotainment unit. The results users who retrieve data act simultaneously as clients and
are from average data communication measurements of servers. It has post-facto gained interest in the research
around 12 MB video streams showing the download time community; see for instance [7]. This system has similarities
when two or five devices are part of the PAN inside the car. to BitTorrent [7], but the mobility assisted delivery means
that data are provided in a random order from a random mix
4.5 PAN Accuracy of peers whereas peer-to-peer content distribution systems
like BitTorrent selects peers based on specific rules. The
When sending data to multiple receivers, a Bluetooth master closest research field is the delay tolerant networking. The
must form a piconet. In a piconet, the master sends data to Delay Tolerant Network Research Group (DTNRG) [14] has
each receiver one packet at a time in a round-robin fashion. proposed architecture [13] to support communication that
To evaluate architecture’s accuracy in piconet mode, we may be used by delay tolerant applications. The architecture
performed the following experiment. We set up a sending consists mainly of the addition of an overlay, called the
device to transmit a very large video file to each slave bundle layer, above a network transport layer. Messages of
joining its piconet. We set up two iPhone as slaves to join any size are transferred in bundles in an atomic fashion that
the piconet one by one, every 90 seconds. We measure the ensures node-to-node reliability. Multicast for delay-tolerant
transfer rate of the initial flow established between the networks has been proposed in [12]. In contrast to multicast,
master and the first joining slave and we plot how this rate our work assumes open user groups. The info-station
changes over time in Figure 7. concept is akin to our proposal and the paper in Ref. [10]
studies means for avoiding exploitation of other nodes. We
differ in that we make the nodal exchanges governed by a
protocol instead of a social contract between users.

In reference [9], it has been shown that delay-tolerant

broadcasting between mobile nodes results in sufficiently
high application level throughput even for streaming. This
is the case in urban pedestrian areas with reasonably high
densities of users, as well as in public transportation and in
places where people gather occasionally (e.g., sport fields,
shopping malls, recreational areas). Contact patterns of
human mobility have been analyzed in the Haggle project
[6]. This project aims at developing an application-
independent networking architecture for delay-tolerant
Figure 7. Data communicate rate in PAN scenario, when networks. In contrast, we implement the podcasting service
iPhone and Infotainment Unit has been paired. directly on top of the link layer to exploit application-
specific policies in the way information spreads across the
Our results demonstrate that framework is accurate: its mobile users.
emulate flows behave similarly to Bluetooth flows with
respect to the number of packets exchanged and packet BlueTorrent [8] is a cooperative content sharing system for
sizes. Finally, we found that framework is accurate when Bluetooth. It differs from our approach in the search
running in piconet mode. mechanisms and the content structuring. We rely on a
channel-based content structure with a subscription model
whereas BlueTorrent employs flat structuring with
5. Related Research traditional query string search. TACO-DTN [11] is a
content-based dissemination system for delay tolerant
Bluetooth is a low-cost technology initially designed for networks. It is implemented as a publish/subscribe system
cable replacement [4] but more generally intended for all and was mainly designed to distribute temporal events
kinds of Personal Area Network (PAN) applications [5]. It is whereas our approach is implemented as a pure receiver-
probable that, in the very near future, Bluetooth will be driven system and optimized for dissemination of streaming
embedded in almost every mobile device. These features media. We use bloom filters in the searching for data; a
coupled with the interoperability characteristic provided by survey of the use of bloom filters in networking is given in
Bluetooth specifications [1], make this wireless technology [15].
very appealing for applications in automotive environments
[2]. As an example, Bluetooth headsets are very popular as
wireless audio link to a mobile phone, also for vehicular use. 6. Conclusions
These reasons make Bluetooth the most suited technology
for the design of the low power Wireless Communication We presented, in this paper, architecture of wireless
Network (WCN). opportunistic communication network for In-vehicle
multimedia application. The prototype implementation is
There has been substantial work on peer-to-peer content targeted at devices that communicate over IEEE 802.15. We
distribution for the wireless network and Internet. have evaluated different tradeoffs in the discovery,
BitTorrent is a successful instance of such systems where synchronization, and data communication phases.
 (IJCNS) International Journal of Computer and Network Security, 19
Vol. 2, No. 3, March 2010

The overall resulting performance is promising and shows ACM SIGCOMM CHANTS Workshop, Pisa, Italy,
the feasibility of wireless opportunistic communication September 2006.
inside a vehicle. With the proposed opportunistic [10] W. H. Yuen, R. D. Yates, and S. C. Sung.
communication system, the scope of multimedia content Noncooperative content distribution in mobile
exchange may broaden and that new participatory wireless infostation networks. In Proceedings of the IEEE
broadcasting applications using the proposed concepts will WCNC, 2003.
emerge in the near future. [11] G. Sollazzo, M. Musolesi, and C. Mascolo, “TACO-
DTN: A Time-Aware COntent-based dissemination
We anticipate two main directions for the future work: system for Delay Tolerant Networks,” in Proceedings
of the First International Workshop on Mobile
Ø Develop analytical models for relevant measures Opportunistic Networking, Puerto Rico, June 2007.
of performance for the described ad hoc network. [12] W. Zhao, M. Ammar, and E. Zegura. Multicasting in
Ø Performance analysis of multimedia protocol for delay tolerant networks: Semantic models and routing
algorithms. In Proceedings of the Sigcomm Workshop
different multimedia resources and performance
on Delay Tolerant Networking, August 2005.
attributes.
[13] S. Burleigh, A Hooke, L. Torgerson, K. Fall, V. Cerf,
B. Durst, and K. Scott. Delay-tolerant Networking: An
References Approach to Interplanetary Internet. IEEE
Communications Magazine, 41(6):128–136, 2003.
[14] Delay Tolerant Network Research Group (DTNRG).
[1] Bluetooth Special Interest Group., “Core Specification
http://www.dtnrg.org.
of the Bluetooth System v 3.0”. Apr. 2009.
[15] A. Broder and M. Mitzenmacher. Network applications
[2] R. Nüsser, R. Pelz, “Bluetooth-based Wireless
of bloom filters: A survey. Proceedings of the 40th
Connectivity in an Automotive Environment”, Proc. of
Annual Allerton Conference, 2002.
the IEEE Vehicular Technology Conference, Fall
2000, vol. 4, pp.1935-1942.
[3] S. Burleigh, A Hooke, L. Torgerson, K. Fall, V. Cerf, Authors Profile
B. Durst, and K. Scott. Delay-tolerant Networking: An
Approach to Interplanetary Internet. IEEE Kamal Sharma received the M. Sc. (Electronics) and M. Tech.
Communications Magazine, 41(6):128–136, 2003. (Future Studies & Planning) degrees from Devi Ahilya University
[4] J.C. Haartsen, “The Bluetooth Radio System,” IEEE in 2002 and 2005, respectively. He is currently with Devi Ahilya
Personal Comm., vol. 7, no. 1, pp. 28-36, Feb. 2000. University, Indore, INDIA.
[5] P. Johansson, M. Kazantzidis, R. Kapoor, and M.
Hemant Sharma received the M. Sc., M. Tech. and Ph D
Gerla, “Bluetooth: An Enabler for Personal Area
degrees from Devi Ahilya University in 1996, 1998 and 2009,
Networking,” IEEE Network, vol. 15, no. 5, pp. 28-37, respectively. He is in to designing and developing platform
Sept.-Oct. 2001. architecture for In-vehicle Infotainment Systems for more than 10
[6] Augustin Chaintreau, Pan Hui, Jon Crowcroft, years. He is currently with Delphi Delco Electronics Europe
Christophe Diot, Richard Gass, and James Scott. GmbH, Germany.
Impact of Human Mobility on the Design of
Opportunistic Forwarding Algorithms. In Proceedings Dr. A. K. Ramani received his Master of Engineering (Digital
of IEEE INFOCOM, Barcelona, Spain, April 2006. Systems) and Ph. D, from Devi Ahilya University, Indore. He
[7] M. Izal, G. Urvoy-Keller, E.W. Biersack, P. Felber, A. worked as a research engineer in ISRO Satellite Center, Dept. of
Al Hamra, and L. Garcés-Erice. Dissecting BitTorrent: Space, Bangalore, India, during 1979-83. Since Jan. 1990, he is a
professor with the School of Computer Science at Devi Ahilya
Five Months in a Torrent’s Lifetime. In Proceedings of
University. He was associate professor at University Putra
Passive and Acrive Measurements Conference, April Malaysia, Dept. of Computer Science during May95 toMay99.
2004. During Sept 2005- July 2006, He was with the College of
[8] Sewook Jung, Uichin Lee, Alexander Chang, Dae-Ki Computer Science and Information Technology, at King Faisal
Cho, and Mario Gerla. BlueTorrent: Cooperative University (KFU), Kingdom of Saudi Arabia. He has guided 13
Content Sharing for Bluetooth Users,. In Proceedings PhDs in different areas of Computer Science and Information
of PerCom, NY, USA, March 2007. Technology and has authored about 70 research papers.
[9] Gunnar Karlsson, Vincent Lenders, and Martin May.
Delay-Tolerant Broadcasting. In Proceedings of the
20 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

An Integrated Approach for Legacy Information

System Evolution
Dr. Shahanawaj Ahamad1
1
Department of Computer Science
College of Arts & Science in Wadi Al-Dawasir,
King Saud University
Wadi Al-Dawasir-11991
Kingdom of Saudi Arabia.
[email protected]

Abstract: Future legacy evolution has always been a great This work is motivated by migration approach suggested in
challenge because of continuous changes in business operation [11] that explains why a module based migration approach
influenced by requirements changes, production of large can be implemented for software maintenance, paper also
commercial benefits, information and communication
technologies development. Usually legacies are not been made to
suggested in this direction but with legacy technical aspects
accommodate these fast changing advancements, this is one of and implantation procedures.
the basic challenge of legacy evolution and renovation also
requires forward and backward procedures and specific
knowledge generation for renovators. Web enabling legacy and 3. Modularization
COBOL based applications interaction with e-commerce based Following fig. 1 depicts how the legacy source is divided in
application is potentially hard to maintain and loss huge
to interacting modules.
amount of organizational economical assets, this paper proposes
the solutions procedures in this evolutionary direction so that on
demand legacy evolution can be performed through adaptive
maintenances.

Keywords: Legacy Systems; Software Evolution; Legacy

Modules; Legacy Code; Legacy Restructuring.

1. Introduction
This paper presents a closer look at software renovation and
explains how legacy software evolutions take place for
future change. It is based on the view that an organization's Figure 1: Dividing legacy source in modules
software systems provide valuable functionality that has
been proven in practice. As such, it should be reused Following are some identified issues associated with this
whenever possible. At the same time, the packaging of this procedure:
business functionality is usually far from optimal as they are
often based on old languages, database systems, and • Availability of legacy sources.
transaction monitors, monolithic in design, and non • Language used to develop legacy source.
maintainable as a result of repeated modification without • Complexity of source understanding and
supporting documents. As a consequence, legacy systems comprehension.
are very hard to change. • Status of documentation and complexity in re-
documentation.
A software system can be effectively evolved with following • Implementation of the tools for analysis and results
procedures: description.
• Modularization • Check feasibility of division of modules.
• Restructuring
• Analysis If division of source in modules is feasible and above
mentioned issues are resolved then undertake next sequence
• Reformation of procedural approach.
• Transformation.
Procedures proposed are explained in the following sections.
4. Restructuring
2. Related Work Intercommunication of the modules using organized of
architecture is to replace the internal structure of the legacy
 (IJCNS) International Journal of Computer and Network Security, 21
Vol. 2, No. 3, March 2010

source. Legacy modules are wrapped to interact with the The interactive search for modules can have any one of a
organized architecture, appear as modules. This process is number of different starting points.
only forward process since the only modifications made to
the legacy are: • The first is to interview, if possible, the users,
• Division in sub-modules. maintainers, and designers of the legacy system in order to
• Wrapping of the modules and major parts of the get a picture of the overall functionality of the system, and,
legacy remain untouched. more specifically, to get an impression which functionality
should be preserved and which is redundant.

• Another starting point is to use the persistent data stores.

Of all the different sorts of data playing a role in the legacy
system, it is likely that the data stored in the database
represents business (domain-specific) entities. Following
such data down to the actual computations as program/data
CRUD Analysis which leads to those programs or
procedures that could be candidate (domain-specific)
modules.

• Another starting point consists of the program call

Figure 2: Restructuring procedure
relationships, which is inter-programs relationships
analysis, which can disclose about cohesion and coupling of
The issues associated with this procedure are:
modules, or about the layers built into the legacy system. If
• Reengineer the Data Bases
one procedure invokes many others (high fan-out), and
• Build a software access layer to access the Data
doesn't get invoked itself, it is likely to be a control
Bases (Legacy and New)
(organized) module, with little built-in functionality.
• Restore the programs so that they can access the new
Likewise, if a procedure is called by many others (high fan-
Data Bases.
in), it is likely to be some sort of utility routine, dealing with
• Externalize a wrapper for each module.
error handling or logging. The procedures with both low
• Write organized scripts to interconnect the modules.
fan-in and low fan-out are the ones that are likely to contain
• Test the restructured system.
business logic.

5. Analysis of Legacy Source • The screen /reports sequence can be identified, together
with the key strokes leading to each subsequent
The objective is to examine the sources of the legacy system
screen/reports. Such screen/reports sequences are very close
and extract information from them that reveals their purpose
to use cases, telling what actions an end user performs.
and architecture.
Moreover, following the flow of screen input fields through
the program identifies those program slices that implement
the given use case. This form of analysis can very well be
5.1 Analysis Methods for Legacy Sources
supported by automated, interactive, tools.
The objective of analysis is to extract information from
legacy source that presents their purpose, and architecture.
• Techniques for combining legacy elements into original
Analysis techniques are crucial for before and further
ways in order to arrive at coherent modules include concept
procedures, this section covers those analysis techniques that
analysis and cluster analysis. Such techniques can be used to
can help to find candidate modules for reformation and
spot combined usage of pieces of data and functionality. For
development in a legacy system.
example, they can be used to group data elements into
candidate classes, based on their usage in programs or
Tools collect all classes of information about the system,
procedures which can then be made into methods of the
which is used by the renovation engineer to find candidate
derived class. In particular concept analysis can be used to
modules. There is no single way to find all good modules
display the various combination possibilities in a concise
therefore the tools should provide many different views on
and meaningful manner.
one legacy source system, and show potential combinations
of these views. This requires a hypertext-based browsing
• Search for modules of specific functions, for example, a
mechanism, potentially supported by intelligent agents and
module for valuing insurance options. A hypertext-based
automata producing specialized reports, as well as an on
legacy browsing system can provide various starting points
line explanation mechanism.
for such a search, such as indexes on words occurring in
comments, column names, inferred types, and so on.
22 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

Moreover typical computations necessary for the behavior of • Language knowledge

the module looked may be identified using plan recognition • Domain knowledge.
and these computations may then be packaged into the • System Knowledge.
required modules. • Requirements and strategy.

It is a fact that major parts of the reformation procedure can

6. Reformation
be automated by combining human efforts with full
This is reformation of individual modules and this automation of repetitive tasks as speed, quality,
procedure is to consider the renovation, up gradation of reproducibility, traceability along with human intelligence.
individual modules as depicted in Figure 3.
7.2 Restructuring Support
The issues associated with this procedure are: Division of legacy source resulting from modularization,
• Determine for each module for which renovation now requires replacing the direct connections between
strategy is to use. legacy modules into indirect connections that are leveraged
• Leave as is. by the organized architecture. This procedure is forward
• Completely replace it by commercial off-the- process as the system is reorganized into modules but the
shelve software systems. code of each module is hardly affected. The following
• Perform a very detailed analysis of the component sequences of operations are needed to achieve this:
in order to extract its business logic that can be
used to rebuild or regenerate the modules. • Reengineer the Data Bases.
• Apply the selected reformation strategy. • Build a software access layer to access the Data
• Test the modules. Bases legacy and new.
• Restore the programs so that they can access the new
Data Bases
• Identify the level of granularity at which the legacy
system will be decomposed.

The major issue associated with this are:

• The total number of modules should be manageable.
Figure 3: Reformation with modules • Smaller modules with many cross relationships are
better handled as a single, larger, module.
7. Transformation • Wrap the identified module in order to connect them
with the organized architecture.
The goal is to technically restructure and improve the • Write organized scripts that simulate the
sources of the legacy modules. Transformation is forward connectivity in the original legacy system.
process as the legacy sources are modified.
7.3 Reformation Support
7.1 Transformation Techniques for Legacy Sources The reformation can be started with individual modules.
Transformation techniques perform forward, systematic, This can be originally changes the original code of the
modifications of the legacy module source in order to enable module. The most interesting properties of this approach
their evolution and increase their flexibility. The legacy are:
module source is transformed for:
• Restructure the whole system. • The mutual dependencies between the evolution
• Restructure the code of individual module. projects of the various modules have been
• Apply uniform comment conventions. eliminated or minimized.
• Eliminate deplore language features.
• Convert to a new language version. • The reformation strategy may differ per module as
• Translate to another language. some modules may be replaced by COTS.

The reformation automata have usually three inputs: In those cases that the decision is to perform a detailed
• The sources of the legacy system. reformation based on the existing code modules, typically
• The repository resulting from the system analysis when much useful business logic is contained in its
phase. transformation techniques may be applied to the code of the
• A set of Automata Detailed Requirements module.
specifications for Transformation.
The issues associated with this procedure are:
The crucial elements in successful reformation tools are:
 (IJCNS) International Journal of Computer and Network Security, 23
Vol. 2, No. 3, March 2010

• Transformation intended for code improvement and [6] M. Fowler, K. Beck, J. Brant, W. Opdyke, and D.
applying uniform layout conventions, goto Roberts. Refactoring: Improving the Design of Existing
elimination, code restructuring, and dead code, Code. Addison-Wesley, 1999.
dead data elimination etc. [7] I. Jacobson, M. Griss, and P. Jonsson. Software Reuse;
• Transformation intended for replacement of certain Architecture, Process and Organization for Business
properties of the code and change of the user- Success. Addison Wesley, 1997.
interface or the database engine etc. [8] C. Jones. Applied Software Measurement: Assuring
• Full translation of the code to another language or Productivity and Quality. McGraw-Hill, 1991.
platform and conversion between COBOL dialects [9]Oluwaseyi Adeyinka, “Service Oriented Architecture &
or translation from obsolete 4GL to standard 3GL. Web Services: Guidelines for Migrating from Legacy
Systems and Financial Consideration” Master thesis,
Blekinge Institute of Technology, 2008.
8. Conclusion
[10] M. Simos.Organization domain modelling (ODM)
The work presented in this paper provides the procedural guidebook version 2.0. Technical Report STARS-VC-
integrated approach to overcome the major challenges for A025/001/00, Synquiry Technologies, Inc, 1996. URL:
future business operations which are the alignment with http://www.synquiry.com/. 450 pp.
changing business goals and changing technologies, while [11] B. Meyer and C. Mingins. Component-base
retaining the assets legacy systems which supporting today's development: From buzz to spark. IEEE Computer,
business operations. It is also discussed the automated 23(7):35-37, 1999.
analysis and transformation of legacy systems. The [12] Dr. Vladimir Bacvanski, An Object-Oriented,
necessary techniques for analysis, transformation, and Component-based, Approach to Migrating Legacy
reformation of legacy systems were also discussed. Paper Systems, 2004.
also concludes that the legacy evolution is an economically
motivated task and legacy is a valuable asset for
organization and business operations. Discussed approach Authors Profile
focuses on modularization of whole legacy system, issues
and propose the successive development procedure for the
affective evolution. Dr. Shahanawaj Ahamad is an
active academician and researcher
References in the field of Software Reverse
Engineering with experience of ten
[1] A.V. Aho, R. Sethi, and J.D. Ullman. Compilers. years, working with King Saud
Principles, Techniques and Tools. Addison Wesley, University’s College of Arts and
1986. Science in Wadi Al-Dawasir,
[2] G. Visaggio, Value-based decision model for renovation K.S.A. He is the member of various
processes in software maintenance, Annals of Software national and international academic and research groups, member
Engineering, 9, Kluwer Academic Publishers, May of journal editorial board and reviewer. He is currently working on
2000, pp 215-233. Legacy Systems Migration, Evolution and Reverse Engineering,
[3]G. Visaggio, Ageing of a Data-Intensive Legacy System: published more than twenty papers in his credit in national and
Symptoms and Remedies, Journal of Software international journals and conference proceedings. He holds M.
Maintenance: Research and Practice, John Wiley; vol. Tech. Degree in Information Technology followed by Ph.D. in
13, pp. 281-308, 2001. Computer Science major Software Engineering, supervised many
[4]C. Szyperski. Component Software; Beyond Object- bachelor projects and master thesis.
Oriented Programming. Addison-Wesley, 1998.
[5] E. Stroulia, M. El-Ramly, L. Kong, P. Sorenson, and B.
Matichuck. Reverse engineering legacy interfaces: An
interaction-driven approach. In 6th Working
Conference on Reverse Engineering, WCRE'99, pages
292-301. Society, 1999.
24 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

Improving Information Accessibility In

Maintenance Using Wearable Computers
Lt.Dr.S Santhosh Baboo1 and Nikhil Lobo2
1
Reader
P.G. & Research Dept of Computer Science
D.G.Vaishnav College, Chennai 106
2
Research Scholar
Bharathiar University

Abstract: In aerospace and defense, maintenance is being (Eric Jorgensen, 1994). The information in these manuals is
carried out using technical manuals in hardcopy. Manuals presented in a window frame with navigation tools for
in hardcopy format are very difficult to carry, access zooming and panning allowing easy comprehension.
information and understand while carrying out Information is displayed based on the
maintenance. This paper brings out the concept of
Interactive Electronic Technical Manuals that can run on
Wearable Computers making maintenance more effective
while reducing the effort involved. Manuals are now
compact and easy to handle, easier to access, can be read
in a systematic manner and easily comprehendible.

Keywords: Wearable Computer, Interactive Electronic Technical

Manual, Manuals, Documentation

1. Introduction
An aircraft is required to be maintained in airworthy
condition. During maintenance of aircrafts, a technician is
often required to refer manuals for maintenance procedures. Figure 1. Interactive Electronic Technical Manual
These manuals are at present in hardcopy consuming time
to access the relevant information. These hardcopy manuals
can be replaced by Interactive Electronic Technical Manuals 3. Need for Interactive Electronic Technical
(IETM) and accessed using wearable computers improving Manuals (IETMs) when compared to Paper-
the performance of technicians by increasing accessibility to Based Manuals
information. Moreover these technicians can receive the
latest information updates electronically without waiting for • Information can retrieved easily
the amended manuals in hardcopy. Following are the • Since information is presented in electronic form, less
objectives to be obtained by using Interactive Electronic storage would be required
Technical Manuals instead of paper-based manuals: • As in the case of hardcopies, pages will not be subject to
• 100% identification of causes of problems (fault wear and tear
isolation) • IETMs can be easily loaded into portable and wearable
• Time spent in solving the problem (troubleshooting) computers which maintenance personnel can take to the
decreased by 20 - 25% field
• Error reduction in performing removals and • Difficult procedures can be integrated with multimedia
replacements by 35 - 40% elements like video and animations to make
understanding easier
2. Interactive Electronic Technical Manuals • Information on various configurations of equipments can
(IETMs) be maintained and displayed by storing only the
difference
Interactive Electronic Technical Manuals (IETMs) are
• The IETM can be used as a basis for developing
manuals in electronic format with interactivity designed for
Computer based training Packages on the same subject
display on computers. An IETM is intended to be the
functional equivalent of a paper-based Technical Manual
and in most cases a total replacement for the paper manual
 (IJCNS) International Journal of Computer and Network Security, 25
Vol. 2, No. 3, March 2010

data integrity. Cross-references are dynamic and search

feature is robust.
• The above IETM is integrated with data from other
processes and systems like expert systems, test
equipments and diagnostics

5. Features of Interactive Electronic Technical

Manuals
Authoring Module
To create manuals with their respective chapters,
sections and subsections and incorporate them into the
IETM
Features:
• User Authentication – to enter the Authoring Module a
user is prompted to enter the login identification (ID)
and password
• Manual Description – facility to add or delete manuals
and their respective chapters, sections and subsections
• Builder – an editor that allows users to create individual
pages to be incorporated into the manual
• Publishing – publishing a viewer module

Viewer Module
To view manuals by clicking on their respective
chapters, sections and subsections
Features:
• User Authentication – to enter the Viewer a user is
prompted to enter the login identification (ID) and
password
• Split Window - to view text pages and corresponding
illustrations simultaneously
• Image Viewer - to pan and zoom illustrations
Figure 2. IETMs compared to Paper-Based Manuals • Safety Notes - warnings, cautions and notes used in the
manual to be displayed in separate windows to alert
4. Types of Interactive Electronic Technical users on their importance
Manuals • Acronyms and Abbreviations - Glossary containing the
list of acronyms and abbreviations used in the manuals
• Page images obtained by scanning in rater format with • Find – to search for a specific word in a page
the table of contents, list of tables, list of figures and • Search – to search for a specific word within manuals
index hyperlinked to the respective contents of the • Notes – facility for users to create their own notes while
manual. A user can select a topic from the table of browsing
contents and the respective raster page is displayed. The • History Back and Forward – keeps a track of the order of
page orientation of the manual is retained and it can be pages that have been visited by the user
viewed and directly printed as per format specifications.
• Bookmark- allow a user to bookmark a page to return to
• An ASCII or PDF document in a scrolling text window. it later
In addition to the above hyperlinks it also contains
• Print - to print a desired chapter, section or subsection
hyperlinks to sections, tables and figures. This document
• Help – to guide the user how to use this Viewer Module
can be linked to video, audio and external applications.
May contain raster and vector graphics. Bookmarks,
search and sticky notes are provided. The manual can be
6. Manuals commonly used in IETM
viewed and directly printed as per format specifications. Maintenance Manual
• SGML files with the content structured so that it can be The description and operation of each aircraft system is
viewed as smaller logical blocks of text with very limited explained in this manual with procedures for removal and
use of scrolling. It is viewed as an indexed PDF file. installation of assemblies that constitute that system.
• The information authored is fully structured and Procedure for repair and cleaning is also included along
hierarchical. In a relational database data tagged with with inspection and testing.
SGML is stored to prevent data redundancy and enforce
Illustrated Parts Catalogue
26 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

The assemblies along with their respective parts are A wearable computer is a potential platform for many
mentioned in this manual. Each part has a unique part different applications that require privacy, mobility and
number, nomenclature and the quantity of that part in the continuous access to information (Lehikoinen, 2002).
assembly. Vendor that supplies this part is also mentioned.

Description and Operation Manual

The description and operation of each aircraft system is
explained in this manual.

Flight Manual
The description and operation of each aircraft system is
explained in this manual. Recommended procedures for
normal operations and emergencies are mentioned. Figure 3. WT4000 wearable computer with a scanner
attachment
Consumable Products Manual
For each aircraft system, the list of consumables The WT4000 is 5.7 inches in length, width 3.7 inches
required to carry out maintenance of that system. The part and height of 1.0 inches, weighing approximately 390.2
number, which is unique along with the nomenclature and grams. It has a keyboard consisting of 23 alphanumeric keys
quantity, are mentioned in this manual. and a memory (FLASH/RAM) of 64/128 MB. The display
type is a backlit color TFT having a display resolution of
Master Servicing Schedules QVGA in landscape mode (320x240). The operating system
This manual covers the preventive maintenance is Microsoft Windows CE 5.0 Professional Version. It can
operations to be carried out to maintain the aircraft in be operated in temperatures ranging from -20° to 50° C and
airworthy condition. It also includes basic servicing, can withstand multiple drops on concrete from a height of 4
component replacements and unconditional inspections and feet. Optional accessories include wearable scanners RS309
checks. and RS409.

Storage and Preservation manual 8. Creation of Interactive Electronic Technical

The manual details the procedures to be carried out for Manuals for a Military Helicopter
storage and preservation of that aircraft and its components.
It also includes instructions for packing and transportation Interactive electronic technical manuals were
of an aircraft with its on-board components. developed and implemented on the field using wearable
computers for the following manuals
TTGE • Description and Operation
For maintenance of an aircraft, the tools and ground • Maintenance Manual
support equipment are mentioned in this manual. • Fault Isolation Manual
• Airplane Illustrated Parts Catalog
7. Wearable Computers Following are the metrics collected on the field during
maintenance of the helicopter
A computer that is subsumed into the personal space of
the user, controlled by the wearer, and that is always on and Table 1: Comparison between Paper-Based Manuals and
always accessible (Steve Mann, 1997). Interactive Electronic Technical Manuals for a Military
Helicopter
A portable computer worn around the body of a
technician into which commands can be entered and Task Paper-Based Interactive
executed while the technician is performing maintenance Manuals Electronic
operations. These portable computers normally contain Technical Manuals
voice recognition and head mounted display as input and
output interfaces. The technician is now able to access the
Correct identification of
capabilities of a desktop and is in contact with Interactive causes of 12 problems 7 12
Electronic Technical Manuals containing maintenance (faults)
instructions at all times.
Time spent in solving
Three dominant aspects characterize wearable the 12 problems 82 hours 63 hours
computers: they are always on and always ready, they are (troubleshooting)
totally controlled by the user, and they are considered both
by the user and by the others around to belong to the user’s Number of errors
personal space (Mann, 1997). encountered when 14 9
performing 38 removals
and 12 replacements
The above metrics indicate the following
• Improvement in identification of causes of problems
 (IJCNS) International Journal of Computer and Network Security, 27
Vol. 2, No. 3, March 2010

(fault isolation) from 58% to 100% to handle, easier to access, can be read in a systematic
• Time spent in solving the problem (troubleshooting) manner and easily comprehendible.
decreased by 23%
• Error reduction in performing removals and
replacements by 36%

9. Creation of Interactive Electronic Technical References

Manuals for a Transport Aircraft
[1] Steve Mann, “Smart Clothing: The Wearable
Interactive electronic technical manuals were Computer and WearCam”, Personal Technologies,
developed and implemented on the field using wearable Vol.1 No.1, Mar. 1997.
computers for the following manuals [2] Eric L. Jorgensen, “The Interactive Electronic
• Maintenance Manual Technical Manual Overview – Setting the Stage”,
• Fault Isolation Manual AFEI CALS Expo International, Oct. 1994.
• Airplane Illustrated Parts Catalog [3] Steve Mann, “An Historical Account of the WearComp
and WearCam Inventions Developed for Applications
in 'Personal Imaging”, The First International
Following are the metrics collected on the field during
Symposium on Wearable Computers: Digest of Papers,
maintenance of the transport aircraft
IEEE Computer Society, pp. 66–73, 1997.
[4] Lehikoinen J, “Interacting with wearable computers:
Table 2: Comparison between Paper-Based Manuals and
techniques and their application in wayfinding using
Interactive Electronic Technical Manuals for a Transport
digital maps”. Ph.D. Thesis. University of Tampere.
Aircraft
Department of Computer and Information Sciences,
Report A-2002-2, 2002.
Task Paper-Based Interactive Electronic
Manuals Technical Manuals

Correct identification of 5 9
causes of 9 problems (faults)

Time spent in solving the 9 58 hours 46 hours

problems (troubleshooting)

Number of errors encountered 8 5

when performing 16 removals
and 4 replacements

The above metrics indicate the following

• Improvement in identification of causes of problems
(fault isolation) from 56% to 100%
• Time spent in solving the problem (troubleshooting)
decreased by 21%
• Error reduction in performing removals and
replacements by 38%

10. Conclusion
The metrics collected for maintenance being carried out
on a Military Helicopter and Transport Aircraft using
IETMs on Wearable Computers are analyzed and following
are the benefits
• 100% identification of causes of problems (fault
isolation)
• Time spent in solving the problem (troubleshooting)
decreased by 20 - 25%
• Error reduction in performing removals and
replacements by 35 - 40%
In conclusion Interactive Electronic Technical Manuals
run on Wearable result in manuals being compact and easy
28 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

Remote Laboratory for Teaching Mobile Systems

Adil SAYOUTI1, Adil LEBBAT 1, Hicham MEDROMI1 and Fatima QRICHI ANIBA1
Laboratoire d’Informatique, Systèmes et Énergies Renouvelables
1

Team Architecture of Systems, ENSEM, BP 8118, Oasis, Casablanca, Morocco

[email protected], [email protected], [email protected] and [email protected]

Abstract: This paper presents the current contribution of

ENSEM (Hassan II University) in the Innovative Educational • Remote laboratories (Figure 1), which offer remote
Concepts for Autonomous and Teleoperated Systems project access to real laboratory equipment and instruments;
which aims at create an innovative educational tool to allow • Virtual laboratories (Figure 2), which offer access to a
students to perform remote laboratory experiments on virtual environment using for this simulation software.
autonomous and teleoperated mobile systems. Although the
Internet offers a cheap and readily available communication
channel for teleoperation, there are still many problems that
need to be solved before successful real-world applications can
be realized. These problems include its restricted bandwidth and
arbitrarily large transmission delay, which influence the
performance of the remote control over Internet. In this article,
we propose a solution that consists in equipping the mobile
systems with a high degree of local intelligence in order for
them to autonomously handle the uncertainty in the real world
and also the arbitrary network delay.
Keywords: E-learning, Control Architecture, Software
Architecture, Multi-agents System, Distributed System, Mobile
System, Internet.
Figure 2. Virtual laboratories
1. Introduction
2. Multi-Agent Systems for Autonomous
Experimentation is a very important part of education in
Control
engineering. This is also true for mechatronic engineering,
which is a relatively new field, combining three engineering The organization of a system - or its control architecture -
disciplines: mechanical engineering, electrical engineering determines its capacities to achieve autonomous tasks and to
and software engineering. The equipments needed for react to events [2]. The control architecture of an
experiments in mechatronic are generally expensive. One autonomous mobile system must have both decision-making
solution for expensive equipments is sharing the available and reactive capabilities: situations must be anticipated and
equipments with other universities around the world [1]. the adequate actions decided by the mobile system
Two are the possibilities to realize it: accordingly, tasks must be instantiated and refined at
• Remotely accessible student laboratory facilities - with execution time according to the actual context, and the
the advent of the Internet and its rapidly spreading mobile system must react in a timely fashion to events. This
adoption in almost all spheres of society - have become can be defined as a rational behavior, measured by the
feasible and are increasingly gaining popularity. mobile system's effectiveness and robustness in carrying out
• Virtual reality (VR) is a system which allows one or tasks.
more users to move and react in a computer generated To meet this global requirement, the control system
environment. architecture should have the following properties [3]:
At present, several e-learning laboratories have been • Programmability: a useful mobile system cannot be
developed. It can be distinguish two categories of them: designed for a single environment or task, programmed
in detail. It should be able to achieve multiple tasks
described at an abstract level. The functions should be
easily combined according to the task to be executed.
• Autonomy and adaptability: the mobile system should be
able to carry out its actions and to refine or modify the
task and its own behavior according to the current goal
and execution context as perceived.
• Reactivity: the mobile system has to take into account
events with time bounds compatible with the correct and
efficient achievement of its goals (including its own
Figure 1. Remote laboratories safety). Consistent behavior: the reactions of the mobile
 (IJCNS) International Journal of Computer and Network Security, 29
Vol. 2, No. 3, March 2010

system to events must be guided by the objectives of is 3. System Architecture

task.
The remote laboratories provide a live performance
• Robustness: the control architecture should be able to
laboratory accessible via Internet, which can be used to
exploit the redundancy of the processing functions.
cover the experimental issues in any tele-education system.
Robustness will require the control to be decentralized to
Clearly as bandwidth increases and higher speed network
some extent.
access reaches users; these factors play an important role in
• Extensibility: integration of new functions and definition
user adoption of remote laboratories.
of new tasks should be easy. Learning capabilities are
The concept of remote laboratories is defined as a
important to consider here: the architecture should make
mechatronic workspace for distance collaboration and
learning possible.
experimentation in research or other creative activity, to
We note an interesting link between the desirable properties
generate and deliver results using distributed information
of intelligent control architecture for autonomous mobile
and communication technologies. To implement a remote
systems and the behavior of agent-based systems:
laboratory, a common Internet-based teleoperation model [5]
• Agent-based approaches to software and algorithm
is used as shown in figure3.
development have received a great deal of research
attention in recent years and are becoming widely
utilised in the construction of complex systems.
• Agents use their own localised knowledge for decision-
making, supplementing this with information gained by
communication with other agents.
• Remaining independent of any kind of centralised
control while taking a local view of decisions gives rise
to a tendency for robust behavior.
• The distributed nature of such an approach also provides
a degree of tolerance to faults, both those originating in
the software/hardware system itself and in the wider
environment.
It is for these reasons that we consider an agent-based
system to be a suitable model on which to base an intelligent
control architecture for complex systems requiring a large
degree of autonomy.
Although widely used, multi-agent systems research has also Figure 3. System Architecture
lead to a number of definitions of agency. Once again, in
some cases, these definitions are inconsistent. In our The Remote user, through his Internet navigator, addresses
context, the terms agent or intelligent agent refer to a a http request to a Web server and downloads an application
material or software entity with one or more independent on his work station. A connection is then established
threads of execution, and which is entirely responsible for towards the server in charge of the management of the
its own input and output from/to the environment in which mobile system to control. The user is then able to take the
it is situated [4]. It is therefore autonomous. We assume that remote control of it. In parallel, other connections are also
the agent has well-defined objectives or goals and exercises established towards multi-media servers broadcasting
problem-solving behavior in pursuit of these goals; reacting signals (video, sound) of the system to be controlled.
in a timely fashion. It is this behavior that allows us to refer The Internet network (network without quality of service)
to the agent as intelligent. While being flexible problem limits the quantity of information that can be transmitted
solvers in their own right, the power of agents is only fully (bandwidth) and introduces delays which can make the
realised once multiple agents are combined and remote control difficult or impossible. The solution
communicating. This is referred to as a multi-agent system proposed, through this work, to face the limitations of the
(MAS). As agents are equipped with different abilities and Internet, is founded on the autonomy and the intelligence,
different goals, each agent has a distinct sphere of influence based on multi-agents systems, granted to the mobile system
within the environment in which all the agents are situated. in order to interact with its environment and to collaborate
These spheres of influence may overlap, defining a with the remote user. The need that consists in wanting to
fundamental relationship between agents. Further assign to the mobile system the maximum of autonomy and
relationships may be superimposed through the use of intelligence brought us to examine in the detail the choice of
communication channels. A MAS, therefore, has all the a remote control architecture [6].
basic properties of a complex system: autonomy,
asynchronicity, concurrency, reactivity and extensibility.
30 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

4. Remote Control Architecture The path planner generates the path using as input the goal,
the mobile system localization and the global map of the
3.4 Control architecture environment.
Humans are sophisticated autonomous agents that are able The reactive part of our architecture, based on couples of
to function in complex environments through a combination agents perception / action, allows the mobile system to react
of reactive behavior and deliberative reasoning. Motivated facing the unforeseen events (Figure 5).
by this observation, we propose a hybrid control
architecture, called EAAS [7] for EAS (Equipe Architecture
des Systèmes) Architecture for Autonomous System. Our
architecture combines a behavior-based reactive component
and a logic-based deliberative component. EAAS is useful in
advanced mobile systems that require or can benefit from
highly autonomous operation in unknown environment,
time-varying surroundings, such as in space robotics and
planetary exploration systems, where large distances and
communication infrastructure limitations render human
teleoperation exceedingly difficult.
The proposed generic architecture consists in associating a
deliberative approach for the high part and a reactive
approach for the low part. The deliberative part or
Figure 5. Reactive Part of the EAAS Architecture
hierarchical agent allows decision-making and actions
planning thanks to the use of the agent selection of actions
(Figure 4). This last is composed of three levels: pilot, 3.5 Remote control software architecture
navigator and path planner. A Software architecture has been defined to make remote
control of mobile systems possible. Our software
architecture is based on a set of independent agents running
in parallel.
On the left side of figure 6, the server side is represented. It
is basically composed of three main agents: “Connection
Manager” which manages the different connected clients
according to a Control Algorithm. This one is chosen by the
designer of the system depending on the application:
master/slave, priority, timeout... The “Media” agent
communicates with the camera in order to broadcast signals
(video, images) of the mobile system in its environment.
The “SMA EAAS” (EAS Architecture for Autonomous
Systems) which represents our control architecture. EAAS
architecture is a hybrid control architecture including a
Figure 4. Actions Selection Agent deliberative part (Actions Selection Agent) and a reactive
part. The reactive part is based on direct link between the
The pilot generates the setting points needed for action sensors (Perception Agent) and the effectors (Action Agent).
agent, based on a trajectory provided as an input. This
trajectory is expressed in a different frame (e.g. Cartesian
frame) from that of the setting points. This trajectory
describes, in time, the position, kinematics and/or dynamic
parameters of the mobile system in its workspace. The pilot
function is to convert these trajectories into setting points to
be performed by the action agent. The navigator generates
the trajectories for the pilot based on data received from the
upper level. These input data are of a geometrical type, still
in a Cartesian frame, but not necessarily in the mobile
system frame. Moreover, these data do not integrate
dynamics or kinematics aspects; contrary to the trajectory,
there is not a strict definition of the velocity, the
acceleration or the force versus time. These input data are
called path – continuous or discontinuous – in cartesian Figure 6. Remote control software architecture proposed
frame. The navigator must translate a path into a trajectory.
 (IJCNS) International Journal of Computer and Network Security, 31
Vol. 2, No. 3, March 2010

The right side of the figure represents the client side. Agents standard Internet protocols. First, the communication is
are loaded in a web navigator. The “Remote Client” based on the HTTP protocol, through a home page linking
corresponds to a graphical user interface which allows the the applets used to move the mobile system. These applets
user to send orders to the mobile system and receive are interpreted by the JVM of the browser. Then, the applets
information about the environment. “Sender” and downloaded on the client communicate with the application
“Receiver” agents are used to allow the communication server through the TCP/IP protocol.
between the client and the server. “Pinger” and “Ponger”
agents are used to observe dynamically the network. If the
connexion is accepted, the “Connection Manager” will
inform the “Local Client” agent which achieves the
interface with the “SMA EAAS” to transmit orders
transmission to the mobile system.

5. Application
The mobile system used in our application is a Lego robot.
Lego Mindstorms [8] is a development kit for
manufacturing a robot using Lego blocks, and is gaining
widespread acceptance in the field of technical education.
By using a Mindstorm, a robot can be manufactured for
various purposes and functions. It is beginning to be
considered as a component of experimental equipment in
robotics research.
The Lego mobile robot (Figure 7) is powered by three
reversible motors coupled to wheels and equipped with four
sensors: sonar sensor, sound sensor, light sensor and touch
sensor. The data produced by these sensors are used by Figure 8. Web Interface
perception agent to build a global map of the Lego robot
environment’s. This global map, the goal and the Lego The remote users can pass online tests of knowledge in
robot localisation are used by the actions selection agent to order to follow a formation answering to their needs.
define a plan of actions to achieve its mission. The Lego Different modules of formation (Cursus link) are available
robot is equipped with Bluetooth connection that permits the on our web site, to know: remote control, multi-agents
communication with the application server and facilitates its systems, systems architecture, control architecture and
displacement in the environment in order to reach its autonomous mobile system. We have also set to the remote
objective. users a discussion forum within our application to
interchange their ideas over the remote control subject.

6. Conclusion
In this paper, we have presented a Web-based remote
control application so that Internet users, especially
researchers and students, can control the mobile robot to
explore a dynamic environment remotely from their home
and share this unique robotic system with us. In the first
part, an analysis of the existing control architectures and the
approaches for their development has guided us to design a
hybrid control architecture. It is called EAAS for EAS
Architecture for Autonomous System. The proposed generic
Figure 7. General Architecture architecture consists in associating a deliberative approach
for the high level and a reactive approach for the low level.
The Web interface of our application is designed with the The deliberative level allows decision-making and actions
intention of making the remote control easy for researchers planning thanks to the use of the agent selection of actions.
and students in order to interact with the Lego mobile robot. The reactive level, based on couples of agents perception /
A simple interface is designed to provide as much action, allows the mobile system to react facing the
information as possible for remote control. This user unforeseen events. Then, the software implementation of our
interface consists of several Java Applets as shown in Figure architecture was presented. It is achieved under the shape of
8. It can work on any web browser. a multi-agents system by reason of its autonomy,
The link between client(s) and server is based on two intelligence, flexibility and the various possibilities of
32 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

evolution. In the second part, in order to validate the choice Adil Lebbat received his Degree in High
of our architecture, we presented one of applications Education Deepened in Network & Telecom
in 2004 from the chouaib doukkali
achieved by the system architecture team of the ENSEM.
university, Eljadida, Morocco. In 2006 he
rejoined the system architecture team of the
ENSEM. Her main research is mainly about
References real time distributed platform based on
multi agents systems: Application based on
[1] P. Le parc, J. Vareille and L. Marce, “E-productique ou the core Linux for products of
contrôle et supervision distante de systèmes telecommunications.
mécaniques sur l’Internet”, Journal européen des
systèmes automatisés (JESA), Vol. 38, n° 5, pp. 525-
558, 2004.
[2] C. Novales, G. Mourioux and G. Poisson, “A multi- Hicham Medromi received the PhD in
level architecture controlling robots from autonomy engineering science from the Sophia
to teleoperation”, First National Workshop on Antipolis University in 1996, Nice, France.
He is responsible of the system
Control Architectures of Robots. Montpellier. April
architecture team of the ENSEM Hassan II
6,7 2006. University, Casablanca, Morocco. His
[3] R. Alami, R. Chatila, S.Fleury, M.Ghallab and actual main research interest concern
F.Ingrand, “An architecture for autonomy. The Control Architecture of Mobile Systems
International Journal of Robotics Research”, Special Based on Multi Agents Systems. Since
Issue on Integrated Architectures for Robot Control 2003 he is a full professor for automatic
and Programming, vol. 17, n° 4, pp. 315-337, 1998. productic and computer sciences at the
[4] J. Ferber, “Multi-Agent Systems: An Introduction to ENSEM, Hassan II University, Casablanca.
Distributed Artificial Intelligence”, Addison Wesley
Longman, Harlow, UK, 1999. Fatima Qrichi Aniba received an electrical
Engineer’s degree from the ENSAM in 2003
[5] A. Sayouti, F. Qrichi Aniba and H. Medromi, “Remote
Meknes, Morocco. In 2005 she got her
Control Architecture over Internet Based on Multi Degree in High Education Deepened in
agents Systems”, International Review on Computers automatic productic from the ENSEM,
and Software (I.RE.CO.S), Vol 3, n° 6, pp. 666 – Hassan II University, Casablanca, Morocco.
671, November 2008. In 2005 she rejoined the system architecture
[6] A. Sayouti, “Conception et Réalisation d’une team of the ENSEM. Her main research is
Architecture de Contrôle à Distance Via Internet à mainly about Real Time Architecture Based
Base des Systèmes Multi-Agents”, Phd. Thesis, on Multi Agents Systems.
ENSEM, Hassan II University 2009.
[7] A. Sayouti, H. Medromi, F. Qrichi Aniba, S. Benhadou
and A. Echchahad, “Modeling Autonomous Mobile
System with an Agent Oriented Approach”,
International Journal of Computer Science and
Network Security (IJCSNS), Vol 9, n° 9, pp. 316 –
321,September 2009.
[8] LEGO Company, “Lego mindstorms official page”,
http ://mindstorms.lego.com, 1997.

Authors Profile
Adil Sayouti received the PhD in
computer science from the ENSEM,
Hassan II University in July 2009,
Casablanca, Morocco. In the same year he
received the price of excellence of the best
sustained thesis in 2009. In 2003 he
obtained the Microsoft Certified Systems
Engineer (MCSE). In 2005 he joined the
system architecture team of the ENSEM,
Casablanca, Morocco. His actual main
research interests concern Remote Control over Internet Based on
Multi agents Systems.
 (IJCNS) International Journal of Computer and Network Security, 33
Vol. 2, No. 3, March 2010

Resource Allocation in Wireless Networks

1
Manish Varshney, 2Dr Yashpal Singh, 3Vidushi Gupta
1
Sr Lecturer Deptt of Computer Science & Engg, SRMSWCET Bareilly, India
Email [email protected]
2
Reader Deptt of Computer Science & Engg BIET Jhansi, India
Email [email protected]
3
Lecturer Deptt of Computer Science & Engg, SRMSWCET Bareilly, India
Email [email protected]

Abstract: In this paper, we study about the resource allocation same desired range (2) Spacial reuse: since transmission is
techniques in wireless network ,thus describing utility based focused in a particular direction, the surrounding area in the
functions and various protocols using directional antenna .In other directions can still be used by other nodes to
other words the stud is based on utility-based maximization for communicate ,now coming back to the point of throughput
resource allocation. We consider two types of traffic, i.e., best and fairness required in resource allocation .
effort and hard QoS, and develop some essential theorems for “Throughput” and “fairness,” however, are conflicting
optimal wireless resource allocation. Directional antenna
performance metrics. To maximize system throughput, the
technology provides the capability for considerable increase in
spatial reuse which is essential in the wireless medium. In this
system will allocate more resource to the users in better
paper, a bandwidth reservation protocol for QoS routing in channel conditions. This may cause radio resource
TDMA-based MANETs using directional antennas is presented. monopolized by a small number of users, leading to
The routing algorithm allows a source node to reserve a path to unfairness. On the other hand, to provide fairness to all
a particular destination with the needed bandwidth which is users, the system tends to allocate more resource to the users
represented by the number of slots in the data phase of the in worse channel conditions so as to compensate for their
TDMA frame. The performance of the proposed schemes is shares. As a result, the system throughput may be degraded
evaluated via simulations. The results show that optimal dramatically. The work in [6-7] show that the system can
wireless resource allocation is dependent on traffic types, total behave either throughput-oriented” or “fairness oriented” by
available resource, and channel quality, rather than solely adjusting certain parameters. However, they do not describe
dependent on the channel quality or traffic types as assumed in
how to determine and justify the value of these parameters,
most existing work. Further optimizations to improve the
leaving this trade-off unsolved.
efficiency and resource utilization of the network is provided.
In this paper, we focus on basic techniques required for
Keywords: Utility-based maximization, wireless networks, resource allocation in wireless networks .through the work
resource allocation, , Mobile ad hoc networks (MANETs), quality we came to the basic two factors which are to be resolved ,
of service (QoS), routing, time division multiple access (TDMA). 1)the first factor relates “user satisfaction” for resource
allocation to avoid such a “throughput-fairness” dilemma
Since it is unlikely to fully satisfy the different demands of
all users, we turn to maximize the total degree of user
1. Introduction satisfaction. The degree of user satisfaction with a given
RESOURCE allocation is an important research topic in amount of resource can be described by the utility function
wireless networks [1-7]. In such networks, radio resource is U(r), a non-decreasing function with respect to the given
limited, and the channel quality of each user may vary with amount of resource r. The more the resource is allocated, the
time. Given channel conditions and total amount of more the user is satisfied. The marginal utility function
available resource, the system may allocate resource to users defined by u(r) = dU(r) dr is the derivative of the utility
according to some performance metrics such as throughput function U(r) with respect to the given amount of resource r.
and fairness [1], [2] or according to the types of traffic The exact expression of a utility function may depend on
[3].but another foremost factor that is required in wireless traffic types, and can be obtained by studying the behavior
network is spatial reuse of network. In order to and feeling of users. We leave the work of finding utility
communicate with another node in a particular location, a functions to psychologists and economists, and focus on
node that is transmitting using an omnidirectional antenna maximizing the total utility for a given set of utility
radiates its power equally in all directions. This prevents functions. and 2) the second factor link to the directional
other nodes located in the area covered by the transmission antennas There are different models that are presented in
from using the medium simultaneously. For this purpose the literature for directional antennas [8].In this paper, the
directional antennas are used. Directional antennas allow a multi-beam adaptive array (MBAA) system is used [1]. It is
transmitting node to focus its antenna in a particular capable of forming multiple beams for simultaneous
direction. Similarly, a receiving node can focus its antenna transmissions or receptions of different data messages.
in a particular direction, which leads to increased sensitivity The rest of the paper is organized as follows. In Sec. II,
in that direction and significantly reducing multi-path resource allocation in wireless networks through utility
effects and cochannel interference (CCI). This allows functions are proposed and proved to be optimal under
directional antennas to accomplish two objectives: (1) Power certain conditions while In Sec. III, resource allocation in
saving: a smaller amount of power can be used to cover the wireless networks through directional antennas Finally, the
paper is concluded in Sec. IV.
34 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

2.1.1 Problem Statement and Definitions

2. Resource Allocation In Wireless Networks Suppose that there are n users served by a base station. Let
Through Utility Functions rtotal denote the total amount of radio resource available at
the base station, and r i, the amount of resource to be
allocated to user i. Users with the same kind of traffic may
In [4], a utility-based power control scheme with respect to not feel the same way by given the same amount of resource
channel quality is proposed. In that scheme, users with because the wireless channel quality for each user may not
higher SIR values have higher utilities, and thus are more be identical. Let denote the channel quality 2 of user i, 0 ≤
likely to transmit packets. Therefore, the wireless medium ≤ 1, and i = 1, 2, · · ·n. The smaller the value of the
can be better utilized and the transmission power can be worse the channel quality. Given an amount of resource r i
conserved, which can adapt to channel conditions and and channel quality qi, the amount of resource actually
guarantee the minimum utility requested by each user. In beneficial to user i is given by θ i = · qi . Let T (i) denote the
[8-9], the authors design a utility-based fair allocation type of traffic of user i. The utility function of user I is
scheme to ensure the same utility value for each user. expressed by where (.) is the utility
However, letting users with different traffic demands to function of traffic T (i) and Ui(.) is the utility function for
achieve an identical level of satisfaction may not be an the type of traffic described by UT(i)(.) but taking into
efficient way of using wireless resource. Worse, traffic account the channel quality of user i. The marginal utility
which is difficult to be satisfied tends to consume most of
the system resource leading to another kind of unfairness. In function of Ui(.) is and that of
[5], a utility-based scheduler together with a Forward Error (.) is (.). Our objective is to maximize
Correction (FEC) and an ARQ scheme is proposed. That subject to and ∀ ≥ 0. An
work gives lagging users more resource and thus results in a optimal allocation for n users with total available resource
similar performance level (i.e., fixed utility value) for each is defined as follows. Note that the optimal allocation
user. The work in [13-14] targets at multi-hop wireless may not be unique in the system.
networks Utility functions have also been widely used in
Internet pricing and congestion control [6]. The typical
approach is to set a price to radio resource and to allocate
tokens to users. The objective is then to maximize the
“social welfare” through a bidding process. These kinds of
bidding schemes, while useful for Internet pricing and
congestion control, may not be practical for wireless
networks. In wireless environments, the types of traffic, the
number of users, and channel conditions are all time-
Figure 1. The utility functions of two types of traffic
varying. It would-be very expensive to implement a wireless
• Definition 2.1: A resource allocation R+ = {r 1 ,r 2
bidding process because the users would have to keep
,∙∙∙,r n } for n users is an optimal allocation if for all
exchanging control messages for real-time bidding, and the
feasible allocations Ra= {r ’1,r ’2 ,∙∙∙,r ’n},U(R+ ) ≥
control protocols of the wireless system would also have to
U(Ra ),where U(R+ ) = and U(Ra )=
be modified to accommodate this process. Finally, the
complexity and efficiency of wireless bidding have not been .
analyzed. It is hard to estimate the time elapsed to achieve • Definition 2.2: R = {r 1, r 2 ,∙∙∙, r n } is a full allocation
the Nash equilibrium. We consider two common types of if
traffic: hard QoS and best effort traffic, and propose three
allocation algorithms1 for these two types of traffic, namely, 2.1.2 HQ Allocation for Hard QoS Traffic
1) the HQ allocation for hard-QoS traffic, 2) the elastic Suppose that there are n users in the queue, all with hard
allocation for best effort traffic, and 3) the mixed allocation QoS traffic. Let denote the residual resource in the
for the co-existence of both types of traffic. These three system. The resource allocation algorithm designed for users
allocation schemes are all polynomial time solutions and whose utility functions are all unit-step functions is referred
proved to be optimal under certain conditions, and in any to as the HQ allocation and the output is denoted by RHQ =
case, the difference between the total utilities obtained by {r 1, r 2 ,∙∙∙, r n }. Given the total available resource in the
our solutions and the optimal utility are bounded. The system , the channel quality and utility function
performance of the proposed schemes is validated via (.) for all users i, RHQ can be obtained as follows.
simulations. The results show that optimal wireless resource 1) Initialize ← 0, i = 1, 2, · · · , n; ←
allocation depends on the traffic demand, total available
2) Sort all users i in the queue in descending order of
resource, and wireless channel quality, rather than solely
dependent on channel quality or traffic type as assumed in .
most existing work. 3) Repeat Steps (4) and (5) until the queue becomes empty.
4) Pop out user i who is now at the head of the queue.
2.1 Resource Allocation Using Hard Qos, Best, And 5) If > , then = , ; = −
Mixed Traffic In Wireless Network Through Utility
Based Functions .
 (IJCNS) International Journal of Computer and Network Security, 35
Vol. 2, No. 3, March 2010

The utility function for user i with hard QoS traffic is 4) If the queue is not empty, then pop out the QoS user i at
described by where the head of the queue; else go to Step (8).
is a unit-step function, is the channel quality of this user, 5) For the popped user i: if ≥ , then = ;else
Mi is the kind of QoS traffic, is the preferred amount of
= 0; go to Step (4).
resource to be allocated.
6) Δ = − .
7) If (Δ > 0), then = − ; go to Step (4); else
= 0; go to Step (8);
8) If ( ) < (0), then = ( ));else = 0.
The allocation rule of this mixed allocation is to: 1) allocate
resource to the first k QoS users at the sorted queue, and 2)
then allocate the residual bandwidth (i.e., − ) to
all best effort users based on the elastic allocation. The
value of k is determined based on the requirement that there
is sufficient resource for this QoS user and the utility gain
Figure 2. Allocation ordering of k users in the HQ ΔUk is positive (i.e., − > 0) .
allocation.
2.2 Resource Allocation Using Directional Antennas
2.1.3 . Elastic Allocation for Best Effort Traffic Medium Access Protocols (MAC) protocols for directional
We next consider the best effort traffic. The resource antenna systems can be classified into two categories: on
allocation algorithm for users with concave utility functions demand and scheduled. In the on-demand scheme nodes
is referred to as the elastic allocation and the output is must exchange short signals to establish a communication
denoted by = {r 1, r 2 ,∙∙∙, r n }. Given the total session.
available resource , the channel quality qi and
marginal utility function for each user i, can
be obtained as follows.
1) For each user i, derive , the inverted function of
.
2) Derive , by summing up , over all users i,
i.e, = .
3) Find , the inverted function of .
4) Find , which is equal to ).
5) For all , i = 1, 2, · · · , n,if < (0), then =
; else = 0.
Figure 3. An example of mixed allocation
The allocation rule of this scheme is to 1) derive the
aggregated utility function from the inverse functions of all Data message transmission is done using the
users, 2) calculate the allocated marginal utility from the omnidirectional mode, and reception is done using the
aggregated utility function, and 3) determine for each directional mode. Directional antennas are used to transmit
user. request-to-send (RTS) and receive clear-to-send (CTS)
signals while the receiver antenna remains in the
2.1.4 Mixture of Hard QoS and Best Effort Traffic omnidirectional mode during this exchange. In [13],
Finally, we consider the co-existence of QoS and best effort communicating pairs are set up using the multi-beam
traffic in the system, which is referred to as mixed forming ability of directional antennas. through cashing of
allocation and the output of which is denoted by = {r 1 , the angle of arrival (AoA), Takai [14] avoided the use of the
r 2 , ∙∙∙ , r n }. omnidirectional mode, which is only used when the AoA
Let denote the amount of residual resource to be given to information is not available.
best effort traffic, and ΔUi, the utility gain by allocating
resource to QoS user i. Other notations remain the same as
in the HQ and the elastic allocations. Given the total
available resource , the channel quality and
marginal utility function for each user i, can be
obtained as follows.
1) Initialize ri ← 0, i = 1, 2, · · · , n ; and ← . Figure 4 . (a) Transmission pattern of an omnidirectional
2) Sort all QoS users i in descending order of , and antenna. (b) Transmission pattern of a directional antenna.
store them in the queue.
2.2.1 Directional Antenna System Assumptions And
3) For each best effort user j, derive from ;
Definitions
Find by summing up over all users j; In this paper, it is assumed that each node in the network is
Find , the inverted function of . equipped with an MBAA-antenna system. Each antenna is
capable of transmitting or receiving using any one of k
36 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

beams which can be directed towards the node with which directional antenna systems. Each node keeps track of the
communication is desired. In order for node x to transmit to slot status information of its 1-hop and 2-hop neighbors.
a node y, node x directs one of its k antennas to transmit in This is necessary in order to allocate slots in a way that does
the direction of node y, and node y in turn directs one of its not violate the slot allocation conditions imposed by the
k antennas to receive from the direction of node x. nature of the wireless medium and to take the hidden and
Radio signals transmitted by omnidirectional antennas exposed terminal problems into consideration.
propagate equally in all directions. On the other hand,
directional antennas install multiple antenna elements so 2.2.3 Slot Allocation Conditions For Directional
that individual omnidirectional RF radiations from these Antennas
antenna elements interfere with each other in a constructive A time slot t is considered free to be allocated to send data
or destructive manner. from a node x to a node y if the following conditions are
true:
1) Slot t is not scheduled to receive in node x or scheduled to
send in node y, by any of the antennas of either node (i.e.
antennas of x must not be scheduled to receive and antennas
of y must not be scheduled to transmit, in slot t).
2) Slot t is not scheduled for receiving in any node z, that is
a 1-hop neighbor of x, from node x where y and z are not in
the same angular direction with respect to x (i.e. ∩
Figure 5. Transmission pattern of an MBAA antenna ≠ ).
system with k=4 beams. Each of the k beams can be oriented 3) Slot t is not scheduled for receiving in node y from any
in a different desired direction. The figure shows: (a) Beams node z, that is a 1-hop neighbor of x, where x and z are in
in transmission mode. (b) Beams in reception mode. the same angular direction with respect to y (i.e. ∩ ≠
).
This causes the signal strength to increase in one or multiple 4) Slot t is not scheduled for communication (receiving or
directions. The increase of the signal strength in a desired transmitting) between two nodes z and w, that are 1-hop
direction and the lack of it in other directions are modeled neighbors of x, where w and y are in the same angular
as a lobe. The angle of the directions, relative to the center direction with respect to z (i.e. ∩ ≠ ), and x and z
of the antenna pattern, where the radiated power drops to are in the same angular direction with respect to w (i.e.
one-half the maximum value of the lobe is defined as the ∩ ≠ ).
antenna beamwidth, denoted by β [9]. With the In Figure6, which illustrates allocation rule 2, node x cannot
advancement of silicon and DSP technologies, DSP modules transmit to node y using slot t, because it is already using
in directional antenna systems can form several antenna slot t to transmit to node z, which is in the same angular
patterns in different desired directions (for transmission or direction as node y. In Figure7, which illustrates allocation
reception) simultaneously. Figure 4(a) shows the rule 3, node x cannot allocate slot t for sending to node y
transmission patterns of an omnidirectional antenna. Figure because slot t is already scheduled for sending from node z,
4(b) shows the transmission pattern of a directional antenna that is a 1-hop neighbor of x, and ∩ ≠ . In Figure
.In this paper, it is assumed that an MBAA antenna system 9, which illustrates allocation rule 4, slot t cannot be
is capable of detecting the precise angular position of a allocated to send from x to y because it is already scheduled
single source for locating and tracking neighbor nodes. for communication between two nodes z and w, that are 1-
Figure 5 shows a node equipped with an MBAA antenna hop neighbors of x, where ∩ ≠ and ∩ ≠
array with k=4 beams. Each of the k beams is able to be . When a node S wants to send data to a node D, with a
oriented in a different desired direction. Figure 5(a) shows bandwidth requirement of b slots, it initiates the QoS path
the antenna array in the transmission mode, and Figure 5(b) discovery process
shows the antenna array in the reception mode.

2.2.2 Protocols For Directional Antennas

The networking environment that is assumed in this paper is
TDMA where a single channel is used to communicate Figure 6. Illustration of allocation rule 2.
between nodes. The TDMA frame is composed of a control
phase and a data phase [17]. Each node in the network has a
designated control time slot, which it uses to transmit its
control information. However, the different nodes in the Figure 7. Illustration of allocation rule 3.
network must compete for the use of the data time slots in
the data phase of the frame. In this section, the slot
allocation rules for the TDMA directional antenna
environment are presented. The hidden and exposed
terminal problems make each node’s allocation of slots Figure 8. Illustration of allocation rule 4.
dependent on its 1-hop and 2-hop neighbor’s current use of
2.2.4 The Qos Path Reservation Algorithm
that slot. The model used in this protocol is similar to that
used in [11] and [12], but includes modifications to support
 (IJCNS) International Journal of Computer and Network Security, 37
Vol. 2, No. 3, March 2010

Node S determines if enough slots are available to send the schemes proposed in [18-19] are the examples of the
from itself to at least one of its 1-hop neighbors. If that is ”fairness” scheme (i.e., α = − 1), and the GR+ scheme in [9]
the case, it broadcasts a QREQ(S, D, id, b, x, PATH, NH) to is an example of the ”throughput” scheme. Fig. 8(a)
all of its neighbors. The message contains the following compares the proposed HQ allocation with different
fields: allocation schemes, and Fig. 9(b) compares the proposed
• S, D and id: IDs of the source, destination and the session. elastic allocation with different allocation schemes. Note
The (S,D, id) triple is therefore unique for every QREQ that the axis of in Fig. 9(b) is in the logarithmic scale.
message and is used to prevent looping. Theresults show that the ”throughput-first” scheme has a
• b: Number of slots required. higher total utility when is small, but the ”fixed”
• x: The node ID of the host forwarding this message. allocation one is closer to the proposed scheme as
• PATH: A list of the form (( ). It increases. Finally, when becomes very large, the
contains the accumulated list of hosts and time slots, which ”fairness-first” scheme can achieve the highest utility. In
have been allocated by this QREQ message so far? is the order to verify, and analyze the performance of the
ith host in the path, and is the list of slots used by to directional antennas (protocols presented in paper),
send to . Each of the elements of contains the slot simulation experiments were conducted. Several
number that would be used, along with the corresponding performance measures were computed as the traffic rate
the set of angular groups, , which represents the (messages/second) is varied. The measured parameters are
direction in which the sending antenna of host i must be the overall percentage of packets received successfully, the
pointed, during that slot, to send data to host i + 1. average number of requests per successful acquisition of
• NH: A list of the form (( ). It QoS path , The average number of requests per session, and
contains the next hop information. If node x is forwarding the average QoS path acquisition time
this QREQ message, then NH contains a list of the next hop
host candidates. The couple is the ID of the host,
which can be a next hop in the path, along with a list of the
slots, which can be used to send data from x to is a
list of the slots to be used to send from host i to host i+1
along with the angular group for each slot. has the same
format as in PATH.

3. Performance Analysis
(a)

In this paper we have gone through two methods of resource

allocation i.e resource allocation through utility functions
and the other using directional antennas. Now since both
methods are for resource allocation it is necessary to
analyze the performance of both the above given methods,
analyzing them leads to the comparison of the the resource
allocation method . In this section, we conduct simulations
to evaluate the performance of our allocation algorithms. (b)
firstly analyzing resource allocation in wireless network Figure 9. Utility comparison with different resource
through utility function. So we consider QoS traffic, best allocation schemes a)Qos traffic b)best effort traffic
effort traffic, and the con-existence of both. The simulation
parameters are described as follows. For QoS traffic, the Simulation results clearly demonstrate the increased
utility function is a unit-step function with = 10 and UM efficiency and performance of the network as the number of
= 1, i.e., UQoS(r) = fu(r − 10); for best effort traffic, UBE(r) = directional antennas increases. As was indicated earlier, this
1 − er/10. The value of qi is randomly generated by a uniform increased performance is due to the considerable increase in
distribution over [0, 1]. We then measure the distributions spatial reuse and the ability for each node to simultaneously
of and θ i under different values of . send or receive data in different directions. This functionally
In Fig. 9, different resource allocation schemes are increases the effective number data slots by a multiple of the
compared with the proposed allocation schemes. The number of antennas (or directions) used. This effect
comparison is based on the scheme proposed in [10], which significantly improves performance. As the data shows, the
allocates radio resource proportionally based on factor increase in performance, or speed-up factor, when the
Depending on the setting of the value α, the system can be number of antenna is increased by a factor of 2 (i.e. doubled
tuned to work with different performance metrics. The curve from 1 to 2, and then from 2 to 4) is significant (speed up
denoted ”throughput” is for α = 1, which gives more factor > 1).As expected, however, it still below a theoretical
resources to the users in better channel conditions, thereby speed-up factor of 2. For the first set of experiments for
leading to a larger system throughput. The curve denoted example, the data shows that that ratio of the overall
”fairness” is for α = − 1, giving all users an identical value average percentage (average for all data traffic rates) of
of θi = · qi. The curve denoted ”fixed” is for α = 0, which successful packets of the two-antenna case to the one-
provides the same amount of resource to all users. Note that antenna case is 1.61, which is > 1 and < 2. The ratio for the
38 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

four-antenna case to the two antenna case is 1.84, which is paths due to the extended range of directional antennas
also > 1 and < 2, and the ratio for the four-antenna case to using the same total transmission power compared to the
the one-antenna case is 2.95 which are < 4. This is to be omnidirectional case. In turn this results in reduced end-to-
expected from the theory of parallel and distributed systems end delay. The simulation results clearly show a significant
because the actual speed-up factor is always below the ratio gain in performance with an increase in the number of
of the number of parallel units, or antennas. successfully received packets, as well as a decrease in the
QoS path acquisition time. However, as expected, this gain
in performance is still below the theoretical speed-up factor.
4. Conclusion In the future, we intend to improve this protocol through the
employment of additional optimization techniques. T
his leads us to conclude that existing channel-dependent-
In this paper, we study about two basic method of allocating only resource schemes and schedulers cannot provide
resources in wireless networks the first one Data message optimal allocation in wireless networks. So In addition, we
length:100MB was utility-based maximization for resource intend to perform more simulations in order to further study,
allocation in infrastructure-based wireless networks. analyze and improve the performance of the protocol under
different network environments including different mobility
rates, and traffic conditions.

References
[1] D.Angelini and M. Zorzi, “On the throughput and
fairness performance of heterogeneous downlink
packet traffic in a locally centralized CDMA/TDD
system,” in Proc. IEEE VTC-Fall 2002.
(a) [2] Furuskar et al., “Performance of WCDMA high speed
Packet data,” in Proc. IEEE VTC 2002-Spring.
[3] Y. Cao, V. O. K. Li, and Z. Cao, “Scheduling delay-
sensitive and besteffort traffic in wireless networks,”
in Proc. IEEE ICC 2003.
[4] M. Xiao, N. B. Shroff, and E. K. P. Chong, “A utility-
based power control scheme in wireless cellular
systems,” IEEE/ACM Trans. Netw., vol. 11, no. 2,
2003.
(b) [5] X. Gao, T. Nandagopal, and V. Bharghavan,
Figure 10. Simulation results. “Achieving application level fairness through utility-
based wireless fair scheduling,” in Proc. IEEE
In this we develop some essential theorems for utility- Globecom 2001.
based resource management. Then, three polynomial time [6] F. P. Kelly, “Charging and rate control for elastic
resource allocation algorithms are proposed for two types of traffic,” European Trans. Telecommun., Jan. 1997.
utility functions. We prove that, in any case, the difference [7] V. A. Siris, B. Briscoe, and D. Songhurst, “Economic
between the total utilities obtained by our proposed models for resource control in wireless networks,” in
solutions and the optimal utility is bounded, and under Proc. IEEE PIMRC 2002, Lisbon, Portugal, Sept. 2002.
certain conditions, all these three schemes can achieve the [8] L. Chen, S. H. Low, and J. C. Doyle, “Joint congestion
maximum total utility (i.e., optimal). From the simulation control and media access control design for wireless ad
results, we find that different types of traffic require hoc networks,” in Proc. IEEE Infocom, Miami, FL,
different kinds of schemes to achieve optimal allocation. In March 2005.
addition, when is small, the system tends to allocate [9] L. Bao and J. J. Garcia-Luna-Aceves. Transmission
more resources to the users in better channel conditions, i.e., scheduling in ad hoc networks with directional
”throughput-oriented;” however, when is abundant, the antennas. Proc. of the 8th annual international
system becomes ”fairness-oriented,” meaning that even with conference on mobile computing and networking, pages
the same traffic, the preference tendency between 48–58, September 2002.
throughput and fairness can still differ. The second one was [10] Q. Dai and J. Wu. Construction of power efficient
a protocol for TDMA-based bandwidth reservation for QoS routing tree for adhoc wireless networks using
routing in MANETs using directional antennas The protocol directional antenna. Distributed Computing Systems
takes advantage of the significant increase in spacial reuse Workshops. Proceedings. 24th International
provided by the directional antenna environment, which Conference on, pages 718–722, March 2004.
drastically increases the efficiency of communication in [11] Jawhar and J. Wu. A race-free bandwidth
MANETs. This is due to the reduction in signal reservation protocol for QoS routing in mobile ad
interference, and the amount of power necessary to establish hoc networks. Proc. of the 37th Annual Hawaii
and maintain communication sessions. Additionally, this International Conference on System Sciences
protocol provides for a relatively smaller hop count for QoS (HICSS’04), IEEE Computer Society, 9, January 2004.
 (IJCNS) International Journal of Computer and Network Security, 39
Vol. 2, No. 3, March 2010

[12] W.-H. Liao, Y.-C. Tseng, and K.-P. Shih. A TDMA- Algorithms, Compiler Design books for the technical students of
based bandwidth reservation protocol for QoS routing graduation and postgraduation.He has published various research
in a wireless mobile ad hoc network. Communications, papers in National and International journals. He has also
ICC 2002. IEEE International Conference on, attended one faculty development program organized by Oracle
Mumbai on Introduction to Oracle 9i SQL and DBA
5:3186–3190, 2002.
Fundamental.
[13] S.; You J.; Hiromoto R.E.; Nasipuri, A.; Ye. A MAC
protocol for mobile ad hoc networks using directional Vidushi Gupta received her B.tech (C.S)
antennas. Wireless Communications and Networking degree from Uttar Pradesh Technical
Conference, 2000. WCNC. 2000 IEEE, 3(23-28):1214– University, Lucknow.She is also pursuing
1219, September 2000. M.tech from Karnataka University, She is
[14] M. Takai, J. Martin, and R. Bagrodia. Directional working as Lecturer ( CS/IT department) in
virtual carrier sensing for directional antennas in SRMSWCET, and Bareilly She has also
mobile ad hoc networks. In Proc. ACM International attended one faculty development program based on the
“Research Methodologies”.
Symposium on Mobile Ad Hoc Networking and
Computing (MOBIHOC), Lausanne, Switzerland, June
2002.
[15] J. Ward and R.T.Jr. Compton. High throughput slotted
aloha packet radio networks with adaptive arrays.
Communications, IEEE Transactions on, 41(3):460–
470, March 1993.
[16] J. Zander. Slotted aloha multihop packet radio networks
with directional antennas. Electronics Letters,
26(25):2098–2100, December 1990.
[17] C. R. Lin and J.-S. Liu. QoS routing in ad hoc wireless
networks. IEEE Journal on selected areas in
communications, 17(8):1426–1438, August 1999.
[18] Jawhar and J. Wu. Qos support in TDMA-based mobile
ad hoc networks. The Journal of Computer Science and
Technology (JCST), Springer, 20(6):797–810,
November 2005.
[19] Jawhar and J. Wu. Race-free resource allocation for
QoS support in wireless networks. Ad Hoc and Sensor
Wireless Networks: An International Journal, 1(3):179–
206, May 2005

Authors Profile
Dr. Yahpal Singh is a Reader and HOD (CS)
in BIET, Jhansi (U.P.). He obtained Ph.D.
degree in Computer Science from
Bundelkhand University, Jhansi. He has
experience of teaching in various courses at
undergraduate and postgraduate level since
1999. His areas of interest are Computer
Network, OOPS, DBMS. He has authored
many popular books of Computer Science for graduate and
postgraduate level. He has attended many national and
international repute seminars and conferences. He has also
authored many research papers of international repute.

Manish Varshney received his M.Sc (C.S)

degree from Dr. B.R.A. University, Agra,
M.Tech. (IT) from Allahabad University and
Pursuing PhD in Computer Science. He is
working as a HOD (CS/IT) in SRMSWCET
Bareilly. He has been teaching various
subjects of computer science for more than
half a decade. He is known for his skills at
bringing advanced computer topics down to the novice's level. He
has experience of industry as well as teaching various courses. He
has authored various popular books such as Data Structure,
Database Management System, Design and Implementation of
40 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

Comparative Analysis of Wireless MAC Protocol

for Collision Detection and Collision Avoidance in
Mobile Ad Hoc Networks
D.Sivaganesan1, Dr.R.Venkatesan2
1
Department of Computer Science and Engineering,
Karpagam College of Engineering, Coimbatore-32, TamilNadu, India.
[email protected]
2
Department of Information Technology.
PSG College of Technology, Coimbatore, TamilNadu, India.

collision detection requires a node to transmit and listen at

Abstract: Packet collisions at the Medium Access Control
(MAC) layer in distributed wireless networks use a combination the same time for terminating a possible collision. Although
of carrier sensing and collision avoidance. When the collision CSMA/CD have been proven to be very successful in wired
avoidance strategy fails, such schemes cannot detect collisions LANs, it cannot be directly employed in wireless networks
and corrupted data frames are still transmitted in their entirety, because of two problems [21]. The first is the hidden
thereby wasting the channel bandwidth and significantly terminal problem [3]. Two mutually hidden terminals are
reducing the network throughput. To address this problem, this two nodes that cannot sense each other (due to the distance
paper compares the wireless MAC protocol CSMA, MACA and
or obstacles between them) but can still interfere with each
IEEE802.11 capable of collision detection and collision
avoidance. The performance of the MAC protocol has been other at a receiver. With hidden terminals, carrier sense
investigated using extensive analysis and simulations. Our alone cannot effectively avoid collisions. The other problem
results shows that as compared with CSMA, MACA MAC for CSMA/ CD in wireless networks is that, in the same
protocols, the protocol IEEE802.11 has significant performance wireless channel, the outgoing signal can easily overwhelm
gains in terms of node throughput and reduce the network the incoming signal due to high signal attenuation in
collisions.
wireless channels. This problem makes it difficult for a
Keywords: MAC, collision detection, collision avoidance, sender to directly detect collisions in a wireless channel.
CSMA, CSMA/CD, MACA. Some existing MAC protocols [5],[6],[7],[8] depend on in-
band control frames for exploring the possible future
1. Introduction channel condition for a data frame and also for reserving the
Distributed Medium Access Control (MAC) protocols such medium for the data frame. However, when the collision
as the IEEE 802.11 Distributed Coordination function avoidance strategy fails, a corrupted data frame is still fully
(DCF) are widely used in computer networks to allow users transmitted. Another category of protocols [3],[9],[10]uses
to statistically share a common channel for their data one or more out-of-hand control channels to avoid
transmissions. In wireless networks, a critical drawback of collisions, These protocols are more effective in dealing
distributed MAC protocols is the inability of nodes to detect with hidden terminals and thus, reduce the probability of
collisions while they are transmitting. As a result, collisions in a network. However, they are incapable of
bandwidth is wasted in transmitting corrupted packets, and detecting collisions either, and if the collision prevention
the achieved throughput degrades. This situation is strategies of these protocols fail, then the collided data
exacerbated as the number of nodes in the network frames are still transmitted in their entirety.
increases, since, now, the rate of collisions increases. To
address this issue, this paper presents MAC protocol capable 2. Carrier Sensing and Collision Avoidance
of detecting collisions in wireless networks which The most widely used mechanism to avoid collisions in the
outperforms existing MAC protocols. The Aloha protocol contention-based MAC is probably ‘‘carrier sensing” [11].
[1] was the first MAC protocol proposed for packet radio Which is used in both wired and wireless networks? The
networks. With pure Aloha, a node sends out a packet drawbacks associated with this mechanism that motivates
immediately upon its arrival at the MAC sub layer and a the development of a scheme with collision detection. With
collided packet is retransmitted with a probability P carrier sensing, a node listens before it transmits. If the
immediately or after each packet transmission time. Carrier medium is busy, then the node defers its transmission. After
Sense Multiple Access with Collision Detection the medium has been sensed idle for a specified amount of
(CSMA/CD) [2] employs two mechanisms to enhance the time, the node usually takes a random back off before
medium utilization in aired local area networks transmitting its frame. The random back off is for avoiding
(LANs):“carrier sense” and “collision detection.” Carrier collisions with other nodes that are also contending for the
sense requires a node to listen before transmitting and medium. Besides the “physical” carrier sensing technique
 (IJCNS) International Journal of Computer and Network Security, 41
Vol. 2, No. 3, March 2010

introduced above, the IEEE 802.11 DCF and MACA 3. Spectrum Reuse and the Capture
(Multiple Access Collision Avoidance) also employs a Phenomenon
technique called “virtual” carrier sensing. The virtual
carrier sense technique relies on in-band control frames to The radio spectrum needs to be spatially reused in multihop
deal with hidden terminals. Before sending a data frame in wireless network for improving network through-put. Better
to the idle medium after proper deferrals and back offs, a spectrum reuse allows more transmissions to go on
source sends out a Request to Send (RTS) frame to contact simultaneously in the network without collisions. A
the receiver and reserve the medium a round the source, If phenomenon closely related to spectrum reuse is “capture”
the receiver receives the RTS frame and its channel is which implies that, when two frames collide at a receiver in
determined to be clear, the receiver sends out a Clear to a wireless network, one of the frames may still be correctly
Send (CTS) frame to respond to the sender and reserve the decoded if the received power of the frame is higher than
medium too. The data transmission then begins if the that of the other by a threshold. However, as we now show
handshake and medium reservation process succeeds. the capture effect is not sufficient to eliminate collisions and
Several situations may cause difficulties to the virtual carrier collision detection is required to prevent bandwidth wastage
sensing technique. One of them is the ‘chained’ hidden on corrupted frames. To illustrate the possibility of
terminal phenomenon. Basically, in a data transaction in the collisions in the presence of capture effect, two scenarios are
MAC layer, the CTS frame sent by a receiver to suppress the shown in Figure. 2a and 2b (the nodes are in a line for easy
hidden terminals of the initiating sender may be lost at the demonstration). In the case, nodes A and D are the initiating
receiver’s neighbors due to the receiver’s own hidden senders, whereas nodes B and C are their receivers,
terminals. In such a case, some hidden terminals of the respectively. In the second case, nodes B and C are the
initiating sender may not be suppressed. An example is senders and nodes A and D are their receivers, respectively.
shown in Figure 1, where node A is the initiating sender In these two cases, assuming same transmission power
and node B is the receiver, The CTS frame generated by levels and ambient noise, captures for the data frames may
node B is corrupted at node C (a hidden terminal of node A) easily happen at receivers because the senders are much
by the signals of node D, which is a hidden terminal of node closer to their receivers than the interference sources.
B. Node mobility may also limit the effectiveness of the However, for combating high link error rates,
virtual carrier sensing technique with a small probability. acknowledgments for data frames are widely used in the
With virtual carrier sensing, only nodes that have receive MAC sub layer of wireless networks. Therefore, interference
the medium reservation message know when to defer. may come not only from the initiating senders, but also from
Therefore, when a node newly moves into a neighborhood their receivers. In both cases shown in Figure. 2b, the two
and misses the preceding reservation information, it senders have to finish their transmissions almost at the same
becomes an unsuppressed hidden terminal to an ongoing time for all the data and acknowledgment frames to be
data transaction, another phenomenon that may impact received without collisions. For example, in case A shown
virtual carrier sensing is that the interference range of a in Figure. 2a, if node A finishes its data transmission earlier
node can larger than its data transmission range[12]. than node D, then node B will send its acknowledgment
Therefore, even if a node is out of the range of another node frame to node A while node C is still receiving the data
for successfully receiving its CTS frame, the node may still frame from node D. A collision may therefore easily occur at
interfere with the other node’s data reception. A more node C. Similarly, if node D finishes is transmission earlier,
effective way to suppress hidden terminals is to use an out- then node B may easily have a collision. The same thing is
of-band control channel. With a single data channel, control true for case B. The corrupted frame, however, be an
information cannot be delivered when the data frame is in acknowledgment instead of a data frame. In reality, two
transmission. With an additional control channel, however, nodes may not finish their transmissions the same time,
control signals can always be present whenever necessary, since their frames may have different sizes and their
which improves the ability of hidden terminal suppression. transmissions may begin at different times. Thus, collision
detection is important in these cases to terminate the
colliding transmissions.

Figure 1. Hidden Terminal Problem

Figure 2 a. Exposed Terminal Problem

42 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

Once the node has finished transmitting the frame header in

the data channel, it expects the intended receiver node B to
Case A have received the information and reply with a CTS pulse in
the control channel. The CTS pulse is transmitted by node B
A B during a pause in the pulses being sent by node A in the
control channel. If a node A does not obtain the expected
CTS pulse in the following pause period after the frame
header is transmitted then node A aborts its transmission in
both channels If node A obtains the expected CTS pulse.
C D Then it keeps transmitting. Node A, however may still abort
Case B its transmissions after obtaining the expected CTS pulse if it
detects a pulse of another node in one of its pulse pauses
after, which indicates a colliding situation. If the node
C D aborts its transmissions due to the lack of the expected CTS
pulse or the detection of a pulse of another node, then it
doubles its contention window and then returns to monitor
the control channel. After node A fully transmits the packet,
it expects an acknowledgment from the receiver. If the node
A B
does not obtain the expected acknowledgment, then it
doubles its contention window and starts to monitor the
Figure 2 b. Collision involving capture control channel again to look for a retransmission
opportunity. The whole process repeats until either node A
4. The MAC Protocol IEEE 802.11 obtains an acknowledgment for the packet or the retry limit
is reached. The node discards the packet in the latter case
3.6 IEEE 802.11-Carrier Sensing and Collision and resets its contention window to the minimum size in
Avoidance both cases. The above description is for the case of a unicast
The MAC protocol in this paper assumes that each node has packet. In the case of a broadcast packet, the proposed
the ability to simultaneously transmit on two channels, the protocol uses the basic CSMA protocol as in the IEEE
control and data channels, with two antennas and, their 802.11 DCF. Figure 3. Shows the RTS / CTS dialogue.
associated communication circuitry. The control channel
has a much smaller bandwidth as compared to the data
channel and is used for transmitting medium reservation
related signals, whereas the data channel is for transmitting
the data and acknowledgments. Instead of relying on bit-
based frames, the control channel employs pulses to deliver
control information. The pulses in the control channel are
single-frequency waves with random length pauses. In the
proposed protocol, pulses only appear in the control channel
and the control channel only carries pulses. When a node is
an active sender or receiver in the data channel, it monitors
the control channel all times, except when it itself is
transmitting in the control channel. If a node is transmitting
in the data channel but detects a pulse in the control
channel, then it aborts its transmissions. To describe the
operation of the protocol, we consider what happens when
the MAC sub layer at a node say node A, receives a packet
to transmit to node B. Before node A can transmit, it first
listens to the control channel to make sure that it is idle. If
the control channel is found idle for a period of time longer
than the maximum pause duration of a pulse, then node A
starts a random back off timer whose value is drawn from
the node’s contention window. If the node detects no pulse
before its back off timer expires, then it proceeds to transmit
the packet upon the expiration of its back off timer.
Otherwise, the node cancels It’s back off timer and keeps
monitoring the control channel. As soon as the back off
timer of node A expires, it starts to transmit pulses in the
control channel a long with the packet in the data channel.
 (IJCNS) International Journal of Computer and Network Security, 43
Vol. 2, No. 3, March 2010

therefore terminate their transmissions, and the collision is

resolved automatically. If only one of the two receivers can
correctly read the frame header, then the sender of the other
receiver will, in general, abort its transmissions due to the
lack of a legitimate CTS pulse. The collision on is therefore
also resolved in such a case. If both receivers can correctly
read the frame headers, then each will send back a CTS
pulse with the length specified in the MAC headers of their
respectively received data frames. If the two initiating
senders do not d raw the same CTS length, then the sender
that draws the shorter one may not receive a legitimate CTS
pulse and thus abort its transmissions. If both senders
receive legitimate CTS pulses, then one sender will usually
still need to abort its transmissions (since their
acknowledgment frames may be interfered with or ca use
interference). The collision detection mechanism starts to
work in such a case. With pauses of random lengths, lie
pulses of the two senders will desynchronize each other over
time. A collision is therefore resolved. The above
description of collision detection is not restricted to two
transmitting nodes that are neighbors. Therefore, two nodes
that are hidden terminals to each other still detect each other
if they transmit at the same time.

5. Results
The comparison of MAC protocols CSMA, MACA and
IEEE 802.11 has been simulated using Network Simulator
Glomosim 2.03. In our simulation each node always has
packets to send, and the destination of each packet is
randomly drawn among the neighbors of the node. The
Figure 3. RTS / CTS dialogue IEEE 802.11 MAC results for three wireless MAC protocols in terms of
HANDSHAKE collision rate and throughput. Figure 3 a, 3 b. Shows the
collision rate and throughput simulation graph for CSMA.
3.7 Collision Avoidance and Detection Figure 4 a, 4 b. Shows the collision rate and throughput
This section further explains how the proposed MAC simulation graph for MACA. Figure 5 a, 5 b. Shows the
protocol achieves collision avoidance and collision collision rate and throughput simulation graph for IEEE
detection. As in the CSMA case, the protocol considers it a 802.11.The protocol IEEE802.11 has significant
potential colliding situation when a transmitting node performance gains in terms of node throughput and reduce
detects another transmitting node. For collision avoidance, the network collisions.
the proposed protocol uses handshake and medium
reservation procedures like those used by traditional wireless
MAC protocols. The difference is that, in the proposed
protocol, these procedures are moved to the control channel
where CTS pulses are used for handshaking and the pulse
relay is used for medium reservation. When the collision
mechanism comes into play and this is the essential
difference between the proposed protocol and other wireless
MAC protocols. To understand how the proposed protocol
resolves collisions, we consider the case where two
neighboring nodes cause collisions. If two neighboring
nodes draw the same back off delays at a contention point
for medium access, then they start to transmit signals in the
data and control channels almost at the same time. If both
receivers of the two senders cannot correctly read the frame
headers due to the resulting collision (that is, the address or
another field in the header does not have a legitimate value),
then neither will send back a CTS pulse. Both senders will Figure 3 a. CSMA – Nodes Vs Collision
44 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

Figure 3 b. CSMA – Nodes Vs Throughput Figure 5 a. IEEE 802.11 – Nodes Vs Collision

Figure 4 a. MACA – Nodes Vs Collision

Figure 5 b. IEEE 802.11 – Nodes Vs Throughput

6. Conclusion
From the simulation result, we conclude that, a better
behavior is obtained when using CSMA instead of MACA
because of the RTS / CTS messages. The use of RTS packets
whenever a source has a data packet to send without first
sensing the channel, results in an increase in packet
collisions and hence decreased throughput. The collision
avoidance mechanism incorporated into IEEE 802.11 for the
transmission of RTS packets aids in the reduction of the
number of collision improves the throughput.
Consequently, more data packets reach their destination. We
conclude that, the MAC protocol IEEE802.11 has
Figure 4 b. MACA – Nodes Vs Throughput significant performance gains in terms of node throughput
and reduces the network collisions in Mobile Ad Hoc
Networks.
 (IJCNS) International Journal of Computer and Network Security, 45
Vol. 2, No. 3, March 2010

References [15] J.L. Sobrinho and AS. Krishnakumar, “Real-Time

Traffic over the IEEE 802.11 Medium Access Control
[1] N. Abramson, “The Aloha System- Another Layer, ‘ Bell Labs Technical I., pp. 172-187, 1996.
Alternative for Computer Communications,” Proc. [16] X.Yang and N.H. Vaidya, “Priority Scheduling in
AFIPS Fall Joint Computer Conf.., 1970. Wireless Ad Hoc Networks,” Proc. ACM MobiHoc,
[2] P. Karn, “MACA—A New Channel Access Method June 2002.
for Packet Radio,” Proc. Ninth ARRL Computer [17] S. Singh and C.S. Raghavendra, “PAMAS-Power
Networking Conf, 1990. Aware Multi-Access Protocol with Signaling for Ad
[3] V. Bharghavan, A, Demers, S. Shenker, and L. Hoc Networks,” ACM SIG COMM Computer Comm.
Zhang, “MACAW:A Medium Access Protocol for Rev.. pp. 5-26, 1998.
Wireless LANs,” Proc. ACM Ann,Conf Applications, [18] J.P Monks, V. Bharghavan, and W.W. Hwu, “A
Technologies, Architectures, and Protocols for Power-Controlled Multiple Access Protocol for
Computer Comm.(SIGCOMM ‘94), Aug. 1994. Wireless Packet Networks,” Proc. IEEE INFOCOM.
[4] Z.J. Haas and J. Deng, “Dual Busy Tone Multiple Apr. 2001.
Access (DBTMA)-A Multiple Access Control [19] P.Jacquet, P.Minet, P.Muhlethaler, and N. Rivierre,
Scheme for Ad Hoc Networks,” IEEE Trans. Comm., ‘‘Priority and Collision Detection with Active
vol. 50, pp. 975-985, June 2002. Signaling — The Channel Access Mechanism of
[5] CL. Fullmer and J.J. Garcia-Luna-Aceves, “Solutions HIPERLAN,’’ Wireless Personal Comm., vol. 4, pp.
to Hidden Terminal Problems in Wireless Networks,” 11-25, Jan. 1997.
Proc. ACM Ann. Conf Applications, Technologies, [20] Y.Thy and K. Chua, “A Capacity Analysis for the
Architectures, and Protocols for Computer Comm. IEEE 802.11 MAC Protocol, Wireless Networks,
(SIGCOMM ‘97), Sept. 1997. vol.7, no.2, pp.159-171, Mar, 2001.
[6] C.L. Fullmer and J.J. Garcia-Luna-Aceves, “Floor [21] T. Rappaport, Wireless Communication: Principles
Acquisition Multiple Access (FAMA) for Packet— and Practice .PHI, 1999.
Radio Networks,” Proc. ACM Ann, Conf
Applications, Technologies, Architectures, and
Protocols for Computer Comm.(SIGCOMM ‘95), Authors Profile
Sept. 1995.
[7] F.A, Tobagi and L. Kleinrock, “Packet Switching in D.Sivaganesan received his BE degree in
Radio Channels: Part I—The Hidden Terminal Computer Science Engineering in 1999,
Problem in Carrier Sense Multiple Access and the M.Tech degree in Information Technology in
Busy Tone Solution,” IEEE Trans. Comm.,vol. 23, 2004.He is currently working as an Assistant
pp. 1417-1433, 1975. Professor in the Department of CSE,
[8] IEEE 802.11 Wireless Local Area Networks, Karpagam College of Engineering,
http://grouper.ieee.org/ groups/802/ 11/, 1999. Coimbatore.He is currently pursuing Ph.D.
[9] C. Wu and V.O.K. Li, “Receiver-Initiated Busy-Tone His research interest includes Mobile Computing, Mobile
Multiple Access in Packet Radio Networks,” Proc. Agents, Web Programming, Object Computing, Simulation
ACM Ann. Conf Applications, Technologies, and Microprocessors Based Systems. He has published 10
Architectures, and Protocols for Computer Comm. technical papers in International, National Conferences and
(SIGCOMM ‘87), Aug. 1987. Journals.
[10] Z.J. Haas and J. Deng, “Dual Busy Tone Multiple
Access (DBTMA)-A Multiple Access Control Dr.R.Venkatesan received his PhD
Scheme for Ad Hoc Networks,” IEEE Trans. Comm., degree in Computer Science and
vol. 50, pp. 975-985, June 2002. Engineering. He is currently working as
[11] L. Kleinrock and F.A. Tohagi, “Packet Switching in Professor and Head, Department of
Radio Channels: Part I-Carrier Sense Multiple- Information Technology, PSG College of
Access Modes and Their Throughput-Delay Technology, Coimbatore. His research
Characteristics,” IEEE Trans. Comm., vol. 23, pp. interest includes Simulation and
1400-1416, 1975. Modeling, Software Engineering, Software, Algorithm
[12] K. Xu, M. Gerla, and S. Bae, “How Effective Is the Design, Database Technology, Software Project
IEEE 802.11 RTS/CTS Handshake in Ad Hoc Management, Software Process Management. He has
Networks?” Proc. IEEE Global Telecom, Conf. published 25 technical papers in International, National
(GLOBECOM ‘02), Nov. 2002. Conferences and Journals.
[13] V. Kanodia, C. Li, A. Sabharwal, B. Sadeghi, and
E.Knightly, “Distributed Multi-Hop Scheduling and
Medium Access with Delay and Throughput
Constraints,” Proc.ACM MobiCom, July 2001.
[14] M.Barry, A.T.Campbell, A.Veres, “Distributed
Control Algorithms for Service Differentiation in
Wireless Packet Networks,” Proc. IEEE INFOCOM.
Apr. 2001.
46 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

Data Encryption technique using Location based key

dependent Permutation and circular rotation
Prasad Reddy. P.V.G.D*, K.R.Sudha2 , S.Krishna Rao3

* Department of Computer Science and Systems Engineering, Andhra University, Visakhapatnam, India,
[email protected]
2
Department of Electrical Engineering, Andhra University, Visakhapatnam, India,
[email protected]
3
Department of Computer Science and Systems Engineering, Sir.C.R.Reddy College of Engineering, Eluru,
[email protected]

transfer of credit card information, financial details and

Abstract: Wireless delivers data through public channels other important documents.
to unspecified clients in mobile distributed systems. In The basic goal of most cryptographic system is to transmit
such scenario, a need for Secure Communication arises. some data, termed the plaintext, in such a way that it cannot
Secure communication is possible through encryption of be decoded by unauthorized agents[5][6][7][8][9]. This is
data. A lot of encryption techniques have evolved over done by using a cryptographic key and algorithm to convert
time. However, most of the data encryption techniques are the plaintext into encrypted data or cipher text. Only
location-independent. Data encrypted with such authorized agents should be able to convert the cipher text
techniques can be decrypted anywhere. The encryption back to the plaintext.
technology cannot restrict the location of data decryption. GPS-based encryption (or geo-encryption) is an innovative
GPS-based encryption (or geo-encryption) is an innovative technique that uses GPS-technology to encode location
technique that uses GPS-technology to encode location information into the encryption keys to provide location
information into the encryption keys to provide location based security[12][13]. GPS-based encryption adds another
based security. In this paper a new technique for Data layer of security on top of existing encryption methods by
encryption method using Location based key dependent restricting the decryption of a message to a particular
permutation and circular rotation is proposed for mobile location. It can be used with both fixed and mobile.
information system. The terms location-based encryption or geo-encryption are
used to refer to any method of encryption in which the
Keywords: Location dependent, circular rotation, encrypted information, called cipher text, can be decrypted
Permutation, GPS only at a specified location. If, someone attempts to decrypt
the data at another location, the decryption process fails and
1. Introduction reveals no details about the original plaintext information.
The device performing the decryption determines its
The dominant trend in telecommunications in recent years location using some type of location sensor such as a GPS
is towards mobile communication. The next generation receiver. Location-based encryption can be used to ensure
network will extend today’s voice-only mobile networks to that data cannot be decrypted outside a particular facility -
multi-service networks, able to carry data and video services for example, the headquarters of a government agency or
alongside the traditional voice services. Wireless corporation or an individual's office or home. Alternatively,
communication is the fastest growing segment of it may be used to confine access to a broad geographic
communication industry. Wireless became a commercial region. Time as well as space constraints can be placed on
success in early 1980’s with the introduction of cellular the decryption location.
Adding security to transmissions uses location-based
systems. Today wireless has become a critical business tool
encryption to limit the area inside which the intended
and a part of everyday life in most developed countries.
recipient can decrypt messages. The latitude/longitude
Applications of wireless range from common appliances that
coordinate of node B is used as the key for the data
are used everyday, such as cordless phones, pagers, to high encryption in LDEA. When the target coordinate is
frequency applications such as cellular phones. The determined, using GPS receiver, for data encryption, the
widespread deployment of cellular phones based on the ciphertext can only be decrypted at the expected location. A
frequency reuse principle has clearly indicated the need for toleration distance(TD) is designed to overcome the
mobility and convenience. The concept of mobility in inaccuracy and inconsistent problem of GPS receiver. The
application is not only limited to voice transfer over the sender can also determine the TD and the receiver can
wireless media, but also data transfer in the form of text , decrypt the ciphertext within the range of TD. Denning’s
alpha numeric characters and images which include the model is effective when the sender of a message knows the
 (IJCNS) International Journal of Computer and Network Security, 47
Vol. 2, No. 3, March 2010

recipient’s location L and the time that the recipient will be of the cipher text denoted by X consisting of the
there, and can be applied especially effectively in situations binary bits of the form x11 x12 x13 ……xn 21…..xn28
where the recipient remains stationary in a well-known As the numbers in the first stage of the cipher text are
location. between 0 to 256 we have bits in every number. Here we
The mobile client transmits a target latitude/longitude have 8n 2 binary bits . Thus we divide the string of 8n 2
coordinate and an LDEA key is obtained for data encryption binary bits into 8 substrings each having n 2 binary bits.
to information server. The client can only decrypt the 1. The next stage of the cipher we transpose the
ciphertext when the coordinate acquired form GPS receiver matrix X consisting of 7 substrings and we
matches with the target coordinate. For improved security, a interchange the first bit of the substring with k1th
random key (R-key) is incorporated in addition to the LDEA bit of the entire string. Similarly the second bit
key. In the present paper the objective is to modify the with k2th bit of the entire string. This process is
cipher by introducing the concept of key dependent circular continued for all the bits in X to get the final cipher
rotation. In this the bits are rotated depending upon the R- text.
key after whitening with the LDEA key using the Exclusive 2. The next stage of the cipher we transpose the
– OR operation. matrix X consisting of 7 substrings. The bits in X
2. Random number generator using are rotated left di times where di = ki mod n 2. In the
quadruple vector decryption process the bits are rotated di bits right.
For the generation of the random numbers a quadruple
vector is used[7][10]. The quadruple vector T is generated 3.1 Algorithm
for 44 values i.e for 0-255 ASCII values.
Algorithm for Encryption:
T=[0 0 0 0 0 0 0 0 1 1 ……………… 0 0 0 0 1 1 1 1 2
{
2…………………….. 0 1 2 3 0 1 2 3 0 1
1. read n,K,P,r
……………………..3]
2. For i=1 to n
The recurrence matrix[1][2][3] [4]
{
0 1 0 
3. p=convert(P);
A = 1 1 0 
4. X=p LDEA key
0 0 1 
5. C1=Permute(X)
is used to generate the random sequence for the 0-255 6. C=Lrotate(C1)
ASCII characters by multiplying r=[A] *[T] and }
considering the values to mod 4. The random sequence 7. Write(C)
generated using the formula [40 41 42]*r is generated.[10] }
Algorithm for Decryption:
3. Development of the cipher {
1. read LDEA-key,R-key,n,C
Consider a plain text represented by P which is represented 2. for i=1 to n
in the form P=[Pij] where i=1to n and j=1 to n ---1 {
Let the key matrix be defined by 3. C1=Rrotate(C)
K=[Kij] where i=1 to n and j=1 to n ---2 4. X=permute(C1)
Let the cipher text be denoted by C=[ Cij] where i=1to n and 5. p= X LDEA key
j=1 to n corresponding to the plain text (1) 6. P=convert(p)
For the sake of convenience the matrices P,K and C are }
represented as 7. write P;
P=[p1 p2 ……pn2] }
K=[k1 k2 ……kn2]
C=[c1 c2 ……cn2 ] 4. Illustration of the cipher
Where in the components are taken row wise from the
corresponding matrices. Encryption :
The process of encryption and decryption are shown
1. The components of p are first converted into their The distance between every pair of points in the universe is
corresponding binary bits in the form negligible by virtue of communication facilities. Let us
p11 p12 p13 p14 p15 p16 p17 p18, p21 ……pn2 1…..pn28 reach each point in the sky. This is the wish of scientists.
where p11 p12 .... are the corresponding binary bits
corresponding to p1 p2….. ASCII equivalent
As the numbers in the plain text are between 0 to 256
we have bits in every number. Here we have 8n 2 binary P=[ 84 104 101 32 100 105 115 116 97 110
bits . Thus we divide the string of 8n 2 binary bits into 8 99 101 32 98 101 116 119 101 101 110 32
substrings each having n 2 binary bits. 101 118 101 114 121 32 112 97 105 114 32
2. The plain text P is whitened by using the Exclusive – 111 102 32 112 111 105 110 116 115 32 105
or operation with the LDEA key to get the first stage 110 32 116 104 101 32 117 110 105 118 101
114 115 101 32 105 115 32 110 101 103 108
48 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

105 103 105 98 108 101 32 98 121 32 118 32 102 97 99 105 108 105 116 105 101 115
105 114 116 117 101 32 111 102 32 99 111 46 76 101 116 32 117 115 32 114 101 97
109 109 117 110 105 99 97 116 105 111 110 99 104 32 101 97 99 104 32 112 111 105
32 102 97 99 105 108 105 116 105 101 115 110 116 32 105 110 32 116 104 101 32 115
46 76 101 116 32 117 115 32 114 101 97 107 121 46 84 104 105 115 32 105 115 32
99 104 32 101 97 99 104 32 112 111 105 116 104 101 32 119 105 115 104 32 111 102
110 116 32 105 110 32 116 104 101 32 115 32 115 99 105 101 110 116 105 115 116 115
107 121 46 84 104 105 115 32 105 115 32 46]
116 104 101 32 119 105 115 104 32 111 102
32 115 99 105 101 110 116 105 115 116 115 The distance between every pair of points in the universe is
46] negligible by virtue of communication facilities. Let us
reach each point in the sky. This is the wish of scientists.
LDEA- key
r1=[ 0 0 1 0 0 0 1 1 ]; 5. Cryptanalysis
r2= [1 0 0 1 1 1 1 1 ];
r3= [1 1 1 1 1 1 1 1]; If the latitude and longitude coordinate is simply used as the
key for data encryption; the strength is not strong enough.
X=P xor LDEA key That is the reason why a random key is incorporated into
Permuting with R-key and Rotating right di times where di LDEA algorithm.
= ki mod 16 and transposing Let us consider the cryptanalysis of the cipher. In this cipher
C=[34 88 14 9 32 120 38 3 101 125 86 the length of the key is 8n2 binary bits. Hence the key space
87 53 89 95 85 99 94 110 13 58 120 55 is 28n2 . Due to this fact the cipher cannot be broken by
43 39 104 86 30 33 25 122 69 49 89 111 Brute force attack.
69 102 124 22 27 121 111 119 102 127 30 The Cipher cannot be broken with known plain text attack
91 61 99 31 74 77 60 104 55 50 39 105 as there is no direct relation between the plain text and the
86 94 37 25 122 85 49 89 79 69 100 124 cipher text even if the longitude and latitude details are
22 19 39 104 118 30 35 25 122 77 107 93 known.
78 109 52 122 55 19 109 110 118 54 59 It is noted that the key dependent permutation plays an
27 91 109 123 87 79 109 92 126 53 51 124 important role in displacing the binary bits at various stages
109 55 114 119 31 75 93 103 92 78 29 of iteration, and this induces enormous strength to the
48 121 55 67 113 109 119 70 119 28 91 cipher.
29 65 76 76 4 48 48 19 3 61 105 119
118 103 31 90 93 81 13 73 68 116 36 19 Avalanche Effect
18 61 105 119 118 103 31 90 93 117 93 With change in LDEA key from 2334719 to 2334718 the
79 85 116 125 23 83 101 110 118 22 59 cipher text would change to
25 91 109] G=[34 88 14 9 32 120 38 3 37 121 86
Cipher text C = 87 37 89 94 85 99 94 110 13 58 120 55
"Xx& e}VW5Y_Uc^n:x7+'hV-! zE1YoEf| yowf -[=c 43 103 108 86 30 49 25 123 69 49 89 111
JM<h72'iV^% zU1YOEd| 'hv-# zMk]Nm4z7 mnv6; 69 102 124 22 27 57 107 119 102 111 30
[m{WOm\~53|m7rwK] 90 61 99 31 74 77 60 104 55 50 103 109
g\N 0y7CqmwFw [ ALL 00  =iwvgZ] Q 86 94 53 25 123 85 49 89 79 69 100 124
IDt$ =iwvgZ] u]OUt} Senv ; [m 22 19 103 108 118 30 51 25 123 77 107
Decryption: 93 78 109 52 122 55 19 45 106 118 54
cipher text C= 43 27 90 109 123 87 79 109 92 126 53 51
"Xx& e}VW5Y_Uc^n:x7+'hV-! zE1YoEf| yowf -[=c 60 105 55 114 103 31 74 93 103 92 78 29
JM<h72'iV^% zU1YOEd| 'hv-# zMk]Nm4z7 mnv6; 48 121 55 67 49 105 119 70 103 28 90
[m{WOm\~53|m7rwK] 29 65 76 76 4 48 48 19 3 125 109 119
g\N 0y7CqmwFw [ ALL 00  =iwvgZ] Q 118 119 31 91 93 81 13 73 68 116 36 19
IDt$ =iwvgZ] u]OUt} Senv ; [m 18 125 109 119 118 119 31 91 93 117 93 79
After key dependent permuting and circular rotation the 85 116 125 23 83 37 106 118 22 43 25
ASCII equivalent is 90 109]
The change in the cipher text is 44 bits
P=[ 84 104 101 32 100 105 115 116 97 110
99 101 32 98 101 116 119 101 101 110 32 6. Conclusions
101 118 101 114 121 32 112 97 105 114 32
111 102 32 112 111 105 110 116 115 32 105
In this chapter a cipher is developed using the LDEA key
110 32 116 104 101 32 117 110 105 118 101
dependent permutation and circular rotation as the primary
114 115 101 32 105 115 32 110 101 103 108
concept. The cryptanalysis is discussed which indicates that
105 103 105 98 108 101 32 98 121 32 118
the cipher is strong and cannot be broken by any
105 114 116 117 101 32 111 102 32 99 111
109 109 117 110 105 99 97 116 105 111 110
 (IJCNS) International Journal of Computer and Network Security, 49
Vol. 2, No. 3, March 2010

cryptanalytic attack since this includes transposition of the Authors Profile

binary bits of the plain text at every stage.
Dr Prasad Reddy P V G D, is a Professor of
7. Acknowledgements Computer Engineering with Andhra University,
Visakhapatnam, INDIA. He works in the areas of
This work was supported by grants from the All India enterprise/distributed technologies, XML based
Council for Technical Education (AICTE) project under object models. He is specialized in scalable web
RPS Scheme under file No. F.No.8023/BOR/RID/RPS- applications as an enterprise architect. With over
114/2008-09. 20 Years of experience in filed of IT and teaching,
Dr Prasad Reddy has developed a number of
References products, and completed several industry projects. He is a regular
speaker in many conferences and contributes technical articles to
[1] K.R.Sudha, A.Chandra Sekhar and Prasad international Journals and Magazines with research areas of
Reddy.P.V.G.D “Cryptography protection of digital interest in Software Engineering, Image Processing, Data
signals using some Recurrence relations” IJCSNS Engineering , Communications & Bio informatics
International Journal of Computer Science and
Network Security, VOL.7 No.5, May 2007 pp 203-207
K.R.Sudha received her B.E. degree in
[2] A.P. Stakhov, ”The ‘‘golden’’ matrices and a new kind Electrical Engineering from GITAM, Andhra
of cryptography”, Chaos, Soltions and Fractals 32 ( University 1991.She did her M.E in Power
(2007) pp1138–1146 Systems 1994. She was awarded her Doctorate
[3] A.P. Stakhov. “The golden section and modern harmony in Electrical Engineering in 2006 by Andhra
mathematics. Applications of Fibonacci numbers,” University. During 1994-2006, she worked with
7,Kluwer Academic Publishers; (1998). pp393–99. GITAM Engineering College and presently she
[4] A.P. Stakhov. “The golden section in the measurement is working as Professor in the department of Electrical
theory”. Compute Math Appl; 17(1989):pp613–638. Engineering, Andhra University, Visakhapatnam, India.
[5] Whitfield Diffie And Martin E. Hellman, New
Directions in Cryptography” IEEE Transactions on
Information Theory, Vol. -22, No. 6, November 1976
S.Krishna Rao received his M Tech degree in
,pp 644-654
Computer Science and Systems Engineering
[6] Whitfield Diffie and Martin E. Hellman “Privacy and from Andhra University in 2000.He is
Authentication: An Introduction to Cryptography” presently working as Associate Professor in the
PROCEEDINGS OF THE IEEE, VOL. 67, NO. 3, Department of Computer Science and
MARCH 1979,pp397-427 Engineering in Sir.C.R.Reddy College of
[7] A. V. N. Krishna, S. N. N. Pandit, A. Vinaya Babu “A Engineering, Eluru.
generalized scheme for data encryption technique
using a randomized matrix key” Journal of Discrete
Mathematical Sciences & Cryptography Vol. 10
(2007), No. 1, pp. 73–81
[8] C. E. SHANNON Communication Theory of Secrecy
Systems The material in this paper appeared in a
confidential report “A Mathematical Theory of
Cryptography” dated Sept.1, 1946, which has now
been declassified.
[9] E. Shannon, A Mathematical Theory of
Communication, Bell System Technical Journal 27
(1948) 379–423, 623–656.
[10] A. Chandra Sekhar , ,K.R.Sudha and Prasad
Reddy.P.V.G.D “Data Encryption Technique Using
Random Number Generator” Granular Computing,
2007. GRC 2007. IEEE International Conference, on
2-4 Nov. 2007 Page(s):576 – 576
[11] V. Tolety, Load Reduction in Ad Hoc Networks using
Mobile Servers. Master’s thesis, Colorado School of
Mines, 1999.
[12]L. Scott, D. Denning, Geo-encryption: Using GPS to
Enhance Data Security, GPS World, April 1 2003.
[13] Geo-encryption protocol for mobile networks A. Al-
Fuqaha, O. Al-Ibrahim / Computer Communications
30 (2007) 2510–25
50 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

Implementation of Watermarking for a Blind Image

using Wavelet Tree Quantization
S.M.Ramesh1, Dr.A.Shanmugam2 , B.Gomathy3
1
Senior Lecturer, Dept.of ECE,
Bannari Amman Institute of Technology, Erode- 638401, India
[email protected]
2
Professor, Dept.of ECE,
Bannari Amman Institute of Technology, Erode-638401, India
[email protected]
3
Lecturer, Dept.of CSE,
Bannari Amman Institute of Technology, Erode- 638401, India
[email protected]

mid-frequency to ensure the transparency and robustness of

Abstract: This paper proposes implementation of
watermarking for a blind image using wavelet tree quantization the watermarked image. For suggested inserting the
for copyright protection. In such a quantization scheme, there watermark into the perceptually significant portion of the
exists a large significant difference while embedding a whole DCT transformed image, wherein a predetermined
watermark bit 1 and a watermark bit 0; it then does not require range of low frequency components excludes the DC
any original image or watermark during the watermark component.
extraction. As a result, the watermarked images look lossless in This watermarking scheme has been shown to be robust
comparison with the original ones, and the proposed method can
effectively resist common image processing attacks; especially
against common attacks such as compression, filtering, and
for JPEG compression and low-pass filtering. Moreover, by cropping. The proposed a watermarking method, in
designing an adaptive threshold value in the extraction process, accordance with the multi-threshold wavelet coding
our method is more robust for resisting common attacks such as (MTWC), the successive subband quantization (SSQ) is
median filtering, average filtering, and Gaussian noise. adopted to search for the significant coefficients. The
Experimental results show that the watermarked image looks watermark is added by quantizing the significant coefficient
visually identical to the original, and the watermark can be
in the significant band by using different weights. The
effectively extracted even after either an unintentional image
processing or intentional image attacks. proposed a watermarking method based on the qualified
significant wavelet tree (QSWT). The QSWT is derived
Keywords - Wavelet tree quantization, JPEG compression, from the embedded zerotree wavelet algorithm (EZW). The
Low-pass filtering, Discrete Wavelet Transform.
watermark is embedded in each of the two subbands of the
wavelet tree. Several watermarking methods used two sets of
1. Introduction
coefficients, one to represent the watermark bit 0, and the
The Internet has popularized tremendously fast in our life in other to represent the watermark bit 1. According to the
the last decade. Due to the digitalization of documents, embedded watermark bit, only a set of coefficients are
images, music, videos, etc, people can access and propagate quantized each time. The proposed a blind watermarking
them easily via the network. The watermarking technique approach called differential energy watermarking. A set of
has been widely applied to digital contents for copyright several 8×8 DCT blocks are composed and divided into two
protection, image authentication, proof of ownership, etc. parts to embed a watermark bit. The high frequency DCT
This technique embeds information so that it is not easily coefficients in the JPEG/MPEG stream are selectively
perceptible; that is, the viewer cannot see any information discarded to produce energy differences in the two parts of
embedded in the contents. The issue here is the detection of the same set.
the existence of watermarks in the digital contents to prove The wavelet coefficients of the host image are grouped
ownership. The spatial and spectral domains are two into wavelet trees, and each watermark bit is embedded
common methods for watermarking. In the spatial domain, using two trees. One of the two trees is quantized with
the watermark is embedded in the selected areas on the respect to a quantization index, and both trees exhibit a
texture of the host image. In the spectral domain, since the large statistical difference between the quantized tree and
spread spectrum communication is robust against many the unquantized tree; the difference can later be used for
types of interference and jamming, the host image is watermark extraction. EL [10] improved SHW [11] method
transformed to the frequency domain using methods such as by adding the Human Visual System (HVS) to effectively
Discrete Cosine Transform (DCT) or Discrete Wavelet resist geometric attack. BKL [9] improved SHW [11]
Transform (DWT); then the watermark is embedded in the method by using four trees to represent two watermark bits
 (IJCNS) International Journal of Computer and Network Security, 51
Vol. 2, No. 3, March 2010

to improve visual quality. One of the four trees is quantized difference is greater than the adaptive threshold value;
according to the binary value of the two embedded otherwise, a watermark bit 0 is extracted. Experimental
watermark bits. But these methods [9-11] cannot effectively results show that the proposed method is very efficient for
resist the attacks of low-pass filtering for JPEG resisting various kinds of attacks.
compression. In this paper, propose a blind watermarking
method based on wavelet tree quantization. 2. Watermarking by Quantization of Wavelet
Trees

2.1 Wavelet Trees

A host image of size n by n is transformed into wavelet
coefficients using the 4-level discrete wavelet transform
(DWT).With 4-level decomposition and have 13 frequency
bands as shown in Fig. 1. The parent-child relationship can
be connected between these sub nodes to form a wavelet
tree. If the root consists of more than one node, then an
image will have many wavelet trees after the DWT. In this
case, we have 3 sub bands LH4, HL4, and HH4 as roots, and
the total wavelet trees are 3×(n/24×n/24) after an image of
size n by n is transformed by a 4-level wavelet transform. A
higher resolution level (such as level 3 in Fig.1) has more
significant coefficients than a lower resolution level (such as
level 2 in Fig.1). As we can see, when n = 512, there are 85
Figure 1. The watermark embedding procedure.
coefficients for a wavelet tree constructed from a node in
LH4 to LH1 following the parent-child relationship. For
A watermarking technique is denoted as blind if the
avoiding attacks such as low pass filters, in our proposed
original image is not needed for extraction. In previous
method we only need the largest two coefficients; these two
research, the watermark embedded in the significant
coefficients are selected from one coefficient of LH4 and
coefficients was found to be robust. The common issue is the
four coefficients of the same orientation in the same spatial
use of blind detection to find out whether the extracting
location in LH3 as shown in Fig. 2.
order is the same as the embedding order. Hence, propose a
watermarking method which embeds a watermark bit in the
2.2 The Preprocess
maximum wavelet coefficient of a wavelet tree. The
With a four-level DWT and have 13 frequency bands as
proposed method is different from others, which use two
shown in Fig.1. A higher level subband is more significant
trees or two blocks to embed a watermark bit. We embed the
than a lower subband. Using the LL4 subband as a root is
watermark by scaling the magnitude of the significant
not suitable for embedding a watermark, since it is a low
difference between the two largest wavelet coefficients in a
frequency band that contains important information about
wavelet tree to improve the robustness of the watermarking
an image and easily causes image distortions. Embedding a
watermark in the HH4 subband is also not suitable, since the
subband can easily be eliminated, for example by a lossy
compression. The LH4 subband is more significant than the
HL4 subband, hence the LH4 subband has a higher priority
than the HL4 subband in the selection. A binary watermark
image W comprised of size Nw (≤ S = n/24×n/24) bits is
embedded. We represent each watermark bit as 1 or 0, and
use a 90 pseudorandom function with another seed to shuffle
Nw bits. According to the watermark bits embedded later, we
select Nw non-overlapping wavelet trees and compute the
global average significant difference of the total number of
the Nw wavelet trees using Eq.(1).

Figure 2. The watermark extraction procedure. 1 Nw

ε = { ------ ∑ (maxi – seci)} (1)
The trees are so quantized that they exhibit a sufficiently Nw i = 1
large enough energy difference between a watermark bit 1
and a watermark bit 0, which is then used for watermark Where ε is the global average significant difference in all
extraction. During extraction, an adaptive threshold value is Nw wavelet trees; {⋅} is the floor function; maxi is the local
designed. A watermark bit 1 is extracted if the significant maximum wavelet coefficient of the ith wavelet tree; seci is
52 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

the local second maximum wavelet coefficient of the ith Otherwise, the embed a watermark bit 0 according to Eq. (7)
wavelet tree, 1≤ i ≤ Nw. If the embedded watermark bit is 1, as follows:
the local maximum coefficient is not quantized under if (seci < 0 ), than maxinew = 0, secinew = 0;
Maximum (ε ,T) Δi ≥ ε and is quantized under Δi <
Maximum (ε ,T) . If the embedded watermark bit is 0, the otherwise maxinew = seci (7)
local maximum coefficient is quantized by setting the local
maximum coefficient = local second maximum coefficient. Where secinew denotes the new second maximum coefficient
in the ith wavelet tree after embedding the watermark bit 0.
When embedding a watermark bit 1, the maximum local
2.3 Watermark Embedding
Let maxi and seci be the local maximum wavelet significant coefficient is quantized and added by an
incremental β if Δi < Maximum (ε, T); otherwise (i.e., Δi ≥
coefficient and the local second maximum wavelet
Maximum (ε, T)), it is kept the same as before. The reason
coefficient in a wavelet tree; the difference between both of
for not quantizing the maximum local significant coefficient
them is named as the local significant difference. The select
is that we won’t increase the distortion of the image. Some
a threshold value β as an increment for quantization. The images having small ε imply that their significant difference
larger the β is, the better robustness but the worse distortion is not obvious. We need an extra parameter T to improve the
of the watermarked image will be. Each time at embed a robustness. The larger the T is, the higher probability the
watermark bit and quantize the maximum coefficient in a maxi is quantized to a larger value; but, in the meanwhile,
wavelet tree. the more distortion of the image will be as well. For
example, let ε =12 and Δi = 13. Suppose T is set to be less
if (maxi < 0), then maxi = 0 (2) than ε, such as T =11, maxi will not be quantized as Δi >ε
=12. On the other hand, if T is set to be larger than ε, such
Δi = maxi − seci, (3) as T =14, maxi will be quantized and increased by β as Δi
<T=14. On the contrary, when embedding a watermark bit
Where Δi denotes the significant difference between the 0, the value of maxi is quantized by decreasing to the local
maximum coefficient and the second maximum coefficient second maximum seci and hence the new Δi will be equal to
in the ith wavelet tree of Nw. Because some maximum 0. Based on this strategy, there exists a large 91 energy
coefficients in a wavelet tree may be a negative value, and a difference between embedding watermark bit 1 and
positive value has higher robustness than a negative value watermark bit 0.
under attacks, it will result in Δi more significant if we
modify the maximum coefficient from a negative value to a 3. Design of Watermark Decoder
positive value. When the Δi is more significant, it will be
more accurate at extracting watermark. To achieve the new 3.1 The Decoder Design
maximum coefficient be positive and to decrease the In the proposed method, neither an original image nor an
original watermark image is required for the extraction
distortion of watermarked image due to quantization, the
process. During the embedding process, embed a watermark
new maximum coefficient is set to a smallest positive value
bit 1 by adding an energy β (or β × γ ) to the local maximum
zero here.
wavelet coefficient in the wavelet tree, and embed a
When embed a watermark bit 1, the maxi is quantized by watermark bit 0 by setting maxi = seci Hence, if the
wavelet tree was embedded a watermark bit 0, the local
maxinew = maxi ,if (Δi ≥ Maximum (ε, T)) (4) significant difference between the largest two coefficients
will be close to zero; otherwise, if the wavelet tree was
maxinew = maxi + β, if (Δi < Maximum (ε ,T) embedded a watermark bit 1, the local significant difference
and maxi is root) (5) between the largest two coefficients will be greater than β.
In order to extract watermark bits correctly and the value
maxinew = maxi + β × γ , if (Δi < Maximum(ε ,T) of y by Eq. (8).
and maxi is root) (6)
1 Nw × α
Where maxinew denotes the new maximum coefficient in y = --------- ∑ ϕ j ------ (8)
the ith wavelet tree after embedding the watermark bit. The Nw × α j = 1
maxi located at lower resolution level (child node) is less
robust than those located at higher resolution levels (root Where φ = {max1 - sec1 ,max2 - sec2 ,…..,maxi - seci }, for
node). The reason has been stated in Section II-B. i=1,2,…,N; the sorted φ is denoted as ϕ (φ) = { ϕ1,ϕ 2 ...ϕ
According to the band sensitivity, the coefficients quantized Nw }, ϕ1 < ϕ 2 < .... < ϕ Nw ; α is the scale parameter, 0 <
at different resolution level are given different weights. The α ≤ 1; α is crucial to y and is used to determine how many
quantized coefficient at the lower resolution level is given a percentages of the significant difference in φ can be used for
heavier weight than that at the higher resolution level. the average. Hence, α marks the minimal y value for
Hence, if the maxi does not locate at the highest resolution extracting the watermark. The larger α is, the larger the y
level, we will quantize the maxi by adding γ times of β will be. Suppose all embedded watermark bits are 1 in the
energy, here γ is a scale parameter and we set γ =1.5. watermark. This means that the difference between the
 (IJCNS) International Journal of Computer and Network Security, 53
Vol. 2, No. 3, March 2010

maximum wavelet coefficients and the second maximum The peak signal-to-noise ratio (PSNR) to evaluate the
wavelet coefficients for any embedded wavelet tree is greater quality between the attacked image and the original image.
than β. The value of α should be set as small as possible to After extracting the watermark, the normalized correlation
avoid extraction errors (see Eq. (9),(10)); the reason for this coefficient (NC) is computed using the original watermark
is that it can exclude those big significant differences of the and extracted watermark to judge the existence of the
embedded wavelet trees in Eq. (8). On the other hand, if all watermark. The value of the NC coefficient is defined as
embedded watermark bits are 0 in the watermark, the value follows:
of α must be set as large as possible. Therefore, α is sensitive
to the content of the watermark. 1 w h -1 w w -1
NC = ----------- ∑ ∑ w (i, j) × w ' (i, j) (11)
3.2 Watermark Extraction wh ×ww i = 0 j = 0
Following Eqs. (8),(9) and (10), it would be easy to extract
the watermark. If the local significant difference is greater Where wh and ww are the height and width of the
than or equal to y, where 0 < y ≤ β, then the embedded watermark. w (i, j) and w' (i, j) are the values of the
watermark bit could be 1; otherwise, the embedded coordinate (i, j) in the original watermark and the extracted
watermark bit could be 0. The watermark bit can be watermark, respectively. Here w (i, j) is set to 1 if it is a
extracted based on Eq. (9) and (10) as follows: watermark bit 1; otherwise, it is set to -1. w' (i, j) is set in
the same way. So the value of w(i, j) × w' (i, j) is either -1 or
watermark bit = 1, if (maxi - seci ) > y (9) 1.

watermark bit = 0, otherwise (10) Table 1: Normalized Correlation Coefficients (NC) after
Attacks by JPEG Compression with the Quality Factors
(QF) 10, 20, 30, 40, 50, 60, 70, 80, 90, AND 100.

Normalized
Quality Factors Correlation Extracted
(QF) Coefficients Watermark
(NC)

10 0.33

20 0.65 E

30 0.78 ES

40 0.82 E A
ESHA
(a) (b) 50 0.89 ES

Figure 3. (a) The original image of Lena of size 512×512. 60 0.96 ESH
(b) The original binary watermark of size 32×16.
70 1 ESHA

80 1 ESHA

90 1 ESHA

100 1 ESHA

The Lena (512×512 pixels, 8 bits/pixel) obtained from

attacking the Stir mark benchmark and Photo Impact 11
software tools to simulate common image attacks. While
there is no attack, for the sake of brevity only the Lena
image and the binary watermark are shown in Fig. 3. Fig. 4
ESHA shows the watermarked image and the extracted result. Pre-
(a) (b) determine the scale parameter T at 10, β =7, γ =1.5, and α
=0.7. In the following, consider both geometric and
Figure 4. (a) Watermarked Lena with PSNR = 41.88dB nongeometric attacks. Nongeometric attacks include JPEG
(b) The extracted watermark with NC = 1 compression, low pass filtering, histogram equalization, and
sharpening. From Table 1, the proposed method can
correctly extract the watermark while the quality factors are
4. Experimental Results
greater than 70 and it becomes worse if the quality factor is
54 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

decreased. The less the quality factor is, the more vague the In this paper, propose a wavelet-tree-based blind
extracted watermark is. watermarking method by quantizing the maximum wavelet
coefficient in a wavelet tree. Embed a watermark bit by
quantizing the maximum wavelet coefficient in a wavelet
tree. The trees are so quantized that they exhibit a
4.1 Experiment Analysis sufficiently large enough energy difference between a
Compare the proposed method with BKL [9], EL [10], watermark bit 0 and a watermark bit 1, which difference,
and SHW [11] using the Lena image. The watermark denoted as significant difference, is then useful for later
consists of 256 bit 1and 256 bit 0. The results are shown in watermark extraction. During extraction, an adaptive
Table 2. From Table 2, the PSNR of the proposed method is threshold value is designed. The magnitude of the
better than those of [11]. In this method, it is not so good for significant difference in a wavelet tree is compared to the
the rotation attacks with degree greater than ± 0.70; but it is adaptive threshold value. Furthermore, regarding each
far better than the listed methods; especially for low pass wavelet tree embedded with a watermark bit, we not only
filtering attacks for JPEG compression. can embed more bits in an image but can extract the
watermark without any need of the original image and
Table 2: Comparison of the Proposed Method and the watermark. Moreover by designing an adaptive threshold
Methods in [9], 10] and [11]. value in the extraction process. In addition to the copyright
protection, the proposed method can also be applied to data
hiding and image authentication.
Attacks / SHW EL BKL Proposed
NC PSNR = PSNR= PSNR= Method
38.2dB 40.6dB 41.54dB PSNR =
References
41.88dB
[1] D. P. Mukherjee, S. Maitra, and S. T. Acton, "Spatial
JEPG NA 0.15 0.17 0.33 domain digital watermarking of multimedia objects for
(QF = buyer authentication," IEEE Trans. Multimedia, vol. 6,
10) pp. 1-15, Feb. 2004.
[2] N. Nikolaidis and I. Pitas, "Robust image watermarking
JEPG NA 0.34 0.61 0.65 in the spatial domain," Signal Processing, vol. 66, pp.
(QF = 385-403, 1998.
20)
[3] A. K. Mahmood and A. Selin,"Spatially adaptive
wavelet thresholding for image
JEPG 0.15 0.52 0.79 0.78
(QF = watermarking,"presented at Proc.IEEE ICME, Toronto,
30) 2006 of Conference.
[4] J. R. Hernandez, M. Amado, and F. Perez- Gonzalez,
JEPG 0.23 0.52 0.83 0.82 "DCT- omain watermarking techniques for still images:
(QF = detector performance analysis and a new structure,"
40) IEEE Trans.Image Processing, vol. 9, pp.55- 68, Jan.
2000.
JEPG 0.28 0.52 0.89 0.89 [5] L. Sin-Joo and J. Sung-Hwan, "A survey of
(QF =
watermarking techniques applied to multimedia," in
50)
Proc. IEEE ISIE 2001, pp. 272-277
JEPG 0.46 0.59 0.94 0.96 [6] V. M. Potdar, S. Han, and E. Chang,"A survey of digital
(QF = image watermarking techniques,"presented at Proc.
60) IEEE INDIN 2005 of Conference.
[7] H. J. Wang, P. C. Su, and C. C. J. Kuo, "Wavelet-based
JEPG 0.57 0.63 0.97 1 digital image watermarking," Optics Express,vol.3, pp.
(QF = 491- 496, Dec. 1998.
70) [8] M. Hsieh, D. Tseng, and Y. Huang, "Hiding digital
watermarks using multiresolution wavelet transform,"
JEPG 0.89 0.71 0.99 1 IEEE Trans. Ind. Electron., vol. 48, pp. 875-882, Oct.
(QF =
2001.
80)
[9] B. K. Lien and W. H. Lin,"A watermarking method
JEPG 1 0.78 1 1 based on maximum distance wavelet tree
(QF = quantization," presented at 19th Conf. Computer
90) Vision, Graphics and Image Processing, 2006 of
Conference.
JEPG 1 0.88 1 1 [10] E. Li, H. Liang, and X. Niu,"An integer wavelet based
(QF = multiple logo-watermarking scheme," presented at
100) Proc. IEEE WCICA, 2006 of Conference.
[11] S. H. Wang and Y. P. Lin, "Wavelet tree quantization
5. Conclusion for copyright protection watermarking," IEEE Trans.
 (IJCNS) International Journal of Computer and Network Security, 55
Vol. 2, No. 3, March 2010

Image Processing, vol. 13, pp. 154-165, Feb. 2004. Guide Award” five times from Tamil Nadu state Government. He
[12] G. C. Langelaar and R. L. Lagendijk, "Optimal is also the recipient of “Best Outstanding Fellow Corporate
differential energy watermarking of DCT encoded Member Award” by Institution of Engineers (IE),India -2004 and
images and video,"IEEE Trans. Image Processing, “Jewel of India” Award by International Institute of Education and
Management, New Delhi–2004 and “Bharatiya Vidya Bhavan
vol. 10, pp. 148-158, Jan.2001.
National Award for Best Engineering College Principal 2005” by
Indian Society for Technical Education (ISTE). “Education
[13] F. A. P. Petitcolas, Weakness of existing watermark Excellence Award” by All India Business& Community
Scheme 1997[online] Available: Foundation, New Delhi.
http://www.petitcolas.net/fabien/watermarking/stirma
rk/index.html.
[14] PhotoImpact 11 software, http://www.ulead.com,
Ulead system, Inc.

Authors Profile

S.M.Ramesh received the B.E degree

in Electronics and Communication
Engineering from National Institute of
Technology
(Formerly Regional Engineering
College), Trichy, Bharathidhasan
University, India in the year 2001 and
the M.E, degree in Applied Electronics
from RVS College of Engineering and
Technology, Dindugal, Anna
University, India in the year 2004. From 2004 to 2005, he served
as a Lecturer in the Department of Electronics and Communication
Engineering, Maharaja Engineering College, Coimbatore, India.
From 2005 to 2006, he served as a Lecturer in the Department of
Electronics and Communication Engineering, Nandha Engineering
College, Erode, India. Since June 2006, he served as Senior
Lecturer, in the Department of Electronics and Communication
Engineering, Bannari Amman Institute of Technology,
Sathyamangalam, and Erode, India. He is currently pursuing the
Ph.D. degree in Anna University, Chennai-India, working closely
with Prof. Dr.A.Shanmugam and Prof Dr.R.Harikumar.

Dr.A.Shanmugam received the B.E,

degree in Electronics and
Communication Engineering from PSG
College of Technology., Coimbatore,
Madras University, India in the year
1972 and the M.E, degree in Applied
Electronics from College of
Engineering, Guindy, Chennai, Madras
University, India in the year 1978 and
received the Ph.D. in Computer
Networks from PSG College of
Technology., Coimbatore, Bharathiyar University, India in the year
1994.From 1972 to 1976, he served as a Testing Engineer at Test
and Development Center, Chennai, India. From 1978 to 1979, he
served as a Lecturer in the Department of Electrical Engineering,
Annamalai University, India. From 1979 to 2002, he served
different level as a Lecturer, Asst.Professor, Professor and Head in
the Department of Electronics and Communication Engineering of
PSG College of Technology, Coimbatore, India. Since April
2004.He assumed charge as the Principal, Bannari Amman
Institute of Technology, Sathyamangalam, Erode, India.

He works in field of Optical Networks, broad band computer

networks and wireless networks, Signal processing specializing
particularly in inverse problems, sparse representations, and over-
complete transforms.Dr.A.Shanmugam received “Best Project
56 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

A Hierarchical Genetic Algorithm for Topology

Optimization and Traffic Management in ATM
Networks Utilizing Bandwidth and Throughput
Alaa Sheta1, Malik Braik2 and Mohammad Salamah3
1
Taif University, Information Systems Department,
College of Computers and Information Systems,
Taif, Saudi Arabia
[email protected]
2
Al-Balqa’ Applied University, Information Technology Department,
Prince Abdullah Bin Ghazi Faculty of Science and Information Technology,
Salt, Jordan
[email protected]
3
National Education Center for Robotics (NECR),
King Hussein Foundation,
Amman, Jordan
[email protected]

increasingly used in telecommunications systems for high

Abstract: The Asynchronous Transfer Mode (ATM) network is
expected to become a backbone network for high speed digital speed multimedia services [2], [3].
multimedia services in distributed environments. This paper Recently, control flow and management are core concepts of
explores an optimization based hierarchical approach using the optimal structure of ATM networks; hence, flow control
Genetic Algorithms (GAs) for installing an ATM network with and management have drawn much attention by the
optimal structure and traffic flow. A Hierarchical Genetic researchers in telecommunication field. In addition to that
Algorithm (HGA) is used to solve such an optimization problem. the nodes in the ATM network design must be linked in an
HGA approach can improve the performance of conventional
GA, though it consumes more computation time. Thus, HGA
economical way to handle expected traffic and capacity
approach is capable of reducing the overall delay while constraints [1], [4]. As the models for the design of ATM
increasing the bits transmitted over the network. This will networks are quite complex, and one of the limiting factors
definitely improve the network performance and meets the is the requirement of expensive exchange based equipments,
requirements in multimedia service environments. The therefore, Genetic Algorithms (GAs) were used to solve such
preliminary results indicate that HGA based ATM network problems of traffic assignment and topological design of
design can be very efficacious.
local area networks [5], [6], flow optimization in survivable
Keywords: Genetic Algorithms, ATM Network, Flow ATM networks [7], and flow assignment in ATM networks
Management. [8]–[10]. However, many programming models have been
developed which deals with telecommunication network
planning [11]. In this context, Evolutionary Algorithms
1. Introduction (EAs) have been frequently applied to telecommunication
There is a rapidly demand for telecommunication networks services in the last years [9], [12].
to function efficaciously despite obstacles such as disabling This paper handles an optimal ATM network structure (i.e.
portions of the network, limiting the links capacities, high topology and link capacities) using Hierarchical Genetic
cabling costs, and so on [1]. Thus, networks must constantly Algorithms (HGA) [13], [14]. The objectives of the network
able to maintain an acceptable level of performance. Despite planning are: designing the network structure to carry out
this need, the problem of dynamic redesign of functioning the estimated traffic, and minimizing the cost of network
networks still received a little attention. Today, with the [15],[16]. Consequently, we are dealing with flow
intensifying role of using the Internet and networking assignment and links capacities as complex optimization
technologies, there have been immense advances in using problems using GAs in two levels, the first GA level deals
the telecommunications services in the present information with selection the optimal network structure, and the second
society. Asynchronous Transfer Mode (ATM) is one of the GA level reduces the overall delay while maintain the
most promising networking technologies. ATM was rated to throughput, this cycle will continues until terminated by
reduce the complexity of the network and improve the some condition.
flexibility of traffic performance which has good ability for The remainder of the paper is organized as follows. In
transmitting many kinds of information, and carry traffic Section 2 we explain the details of GA levels; each
over all kinds of networks. Furthermore, ATM has been alternative level of HGA is described extensively. The case
 (IJCNS) International Journal of Computer and Network Security, 57
Vol. 2, No. 3, March 2010

study and the schematic diagram of the ATM network are 2.2 GA Level 2
discussed in section 3. In Section 4 we provide an extensive GA is optimized to find the best value of (fi) such that the
and comprehensive formulation of the use of HGA for overall delay is minimized. The computed network delay is
solving the ATM design problem. In section 5 we present then returned to GA Level 1 to be considered as the new
our preliminary results for the ATM network case study. fittest value. The fitness criterion, in this level, is considered
Finally, in section 6 we offer several concluding remarks and as the mean time delay. The mean delay is computed as in
address the challenges ahead in the future enhancements. Equation 3.

2. Hierarchical GA 1 N fi N
T = ∑ , θ = ∑ fi (3)
The ATM design problem was considered as a parameter θ i =1 C
i − fi i = 1
estimation problem. GAs has been used in a hierarchical
approach to optimize the ATM network. The proposed novel
where θ is the total traffic over the network, N is the number
HGA optimization problem is shown in Figure 1. There are of links. This idea was presented in [16].
two levels were performed to produce an optimal ATM
network design. The two levels are evolved in parallel. The
3. Case Study: ATM Network
fitness score of the individuals in each level depends on the
performance of the chromosomes in both levels. There are two sets of customers to be considered while
planning the ATM network design:
1) the user who uses the services through the network. The
network should meet the user’s needs in terms of quality
of service. The flow assignment and the capacities links
are the two main entities affecting the cost of the ATM
network. The maximum flow assignment must be tested,
and the network should maintain a high capacity.
2) the company which will be building the ATM network
and maintaining it. The network operation should be as
cost effective as possible for both installation and
maintenance. Minimizing the total cost is mainly an
important matter of selecting the best design of ATM
network.
The schematic diagram for the ATM under study with all
specified links and capacities is shown in Figure 2.

Figure 1. Hierarchical GAs in two levels

2.1 GA Level 1
GA is used to install a new network with a set of network
links which satisfies the constraints demand criterion and
has a minimum delay over the network. In this level, a new
network structure is evolved which has all nodes of the
original network and a subset of its links. The network
configuration must be implemented subject to the constraints
defined in Equations 1 and 2.
Figure 2. The ATM Network Topology
f ≤ C × 95% (1)
i i
This ATM network was presented in [17], [18]. The network
Thus, the maximum flow value (fi) for link (i) in Equation 1 has 7 nodes with 14 links. The links and capacities for each
is restricted not to exceed 95% of the capacity (Ci). node are presented in Table 1.
n Each network link is characterized by a set of attributes;
Throughput ≥ 0.5 × ∑ C
i
(2) these attributes for a given link (i) form the flow (fi ) and the
i =1 capacity (Ci ). (fi ) is defined as the effective quantity of
The throughput of the network is allowed not to be less than information transported by link (i). The capacity (Ci ) is
half the sum of all capacities. We must realize that the defined as the measure of the maximal quantity of
boundary range of the genes values is based on the link ID of information that can be transmitted by link (i).
the ATM experiments and the gene redundancy must be
eliminated. A gene duplicate leads to a link duplicate, which
leads to the increase of the network delay.
58 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

Table 1: ATM network capacity of each link selection method is the fitness assignment. Each individual
in the selection pool receives a reproduction probability
Links depending on its objective value. This fitness is used for the
Start Node End Node Link Capacity
ID selection step afterwards. Stochastic uniform selection
1 1 4 150 mechanism is considered in this work.
2 1 6 150 Reproduction: This step increases the number of good
chromosomes in the next generation. This scheme
3 1 3 225
determines how the new individuals will be assimilated into
4 1 2 225
the population. Many number of reproduction operators were
5 2 3 150 presented in the literature [3], [7].
6 2 6 150 Crossover: This procedure exchanges genes between the
7 2 5 150 parents. Two chromosomes are randomly selected from the
8 3 4 150 population as parent chromosomes. Two new chromosomes
with the genes from both the parent chromosomes are
9 3 5 150
obtained. A scattered function for the crossover is considered
10 4 6 150
in this work.
11 6 5 150 Mutation: This step is used to have a new chromosome
12 5 7 150 which differs from the chromosomes in the population. A
13 6 7 150 chromosome is randomly selected as the mutated
14 4 7 150 chromosome. In this paper, the mutation type is Gaussian
function with Scale equal 1 and Shrink equal 1.
Fitness Evaluation: The fitness function is computed by
4. Problem Formulation
using the mean time delay as given in Equation 3. This
The optimization problem of the ATM network is fitness is based on a given objective function on level 2
implemented in a hierarchical manner. Two levels of GAs which is the minimum time delay over the evolved network
are used. in level 1. Each chromosome in the population is assigned a
• GAs Level 1 is used to select the best network topology specific value associated with the gene arrangement in order
and forward it to GA Level 2. The ATM network should to select the best individuals. Finally, the effectiveness of the
be effectively connected such that at least half the proposed algorithm was tested based on the GA utilization.
network bandwidth is busy. The GA utilization criterion can be computed as given in
• GAs level 2 computes the minimum delay over the Equation 4.
network. In GAs level 2 we determine the (fi) parameters f
for each network link (i). Utilization = i (4)
HGA requires number of steps for the problem formulation C
i
as shown below: Improving GA utilization in all cases will increase the flow
Encoding Mechanism: This step is used to encode the assignment of the links for the ATM network. HGA requires
variables of the optimization problem in terms of genes. In number of tuning parameters in accordance with GA
this work, a table was created for all pairs of links optimization. We run GA using various population sizes in
combination. The links in the table correspond to the virtual the two levels of HGA. The corresponding results of the
paths included between pair of nodes. In the proposed network topology for the various population sizes are shown
algorithm each link is identified by an ID number which is in Figures 3, 4, 5, 6, 7, 8, 9, 10 and 11.
in accordance to the row number in the table. The proposed
GAs will have 14 genes. Each gene corresponds to a link
and each link has an identified (Ci).
Initialization: This step uses the encoding method to create
a random initial population by randomly generating a
suitable number of chromosomes. In this paper, the
chromosome length equals to the number of links in the
original network.
Number of Populations: Various population sizes were used
while running GA in each level.
Selection Mechanism: Selection schemes helps in
determining the convergence properties of GAs. A selection Figure 3. Network Topology evolved by Pop. Size L1 = 10
scheme in GAs is known as the process which favorably and Pop. Size L2 = 25
selects better individuals in the current population for the
mating pool [19], [20]. The selection method determines
how individuals are chosen for mating. The first step in the
 (IJCNS) International Journal of Computer and Network Security, 59
Vol. 2, No. 3, March 2010

Figure 8. Network Topology evolved by Pop. Size L1 = 30

Figure 4. Network Topology evolved by Pop. Size L1 = 10 and Pop. Size L2 = 75. Delay time = 10.7585
and Pop. Size L2 = 50. Delay time = 11.0223

Figure 9. Network Topology evolved by Pop. Size L1 = 50

and Pop. Size L2 = 25. Delay time = 10.76630
Figure 5. Network Topology evolved by Pop. Size L1 = 10
and Pop. Size L2 = 75. Delay time = 10.76282

Figure 10. Network Topology evolved by Pop. Size L1 = 50

and Pop. Size L2 = 50. Delay time = 10.7643

Figure 6. Network Topology evolved by Pop. Size L1 = 30

and Pop. Size L2 = 25. Delay time = 10.8009

Figure 11. Network Topology evolved by Pop. Size L1 = 30

Figure 7. Network Topology evolved by Pop. Size L1 = 30 and Pop. Size L2 = 75
and Pop. Size L2 = 50. Delay time = 10.7581
60 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

The HGA related parameters used in the experiments were was presented to show the performance of the proposed
determined as shown in Table 2. HGA algorithm. Figure 12 shows the convergence process of
GAs with various population sizes for both level 1 and level
Table 2: Parameters evolved for GA-based experiments 2.

Operator Type

Creation function random initial with uniform

distribution

Initial range lower bound is 1 and upper bound

is 95% of line capacity

Selection mechanism Stochastic uniform

Crossover type Scattered function

Gaussian function with
Mutation type Scale = 1 and Shrink = 1

Maximum generations 50

5. Experimental Results
From the developed results, the proposed HGA was able to
provide number of ATM network with various topologies.
Having a closer look to each developed structure, we can Figure 12. The convergence process of GAs with various
find that all evolved network have seven links except one population sizes
network which has 8 links (see Figure 4). This network has
the maximum delay which is expected since the number of The algorithm converges to the optimal solution at ”high”
links increased. The evolved network were able to achieve speed and finds a good solution in less than 20 generations
the required criteria which is specified by the designer (i.e. of running. Moreover, the algorithm reached a solution with
increasing the network throughput to at least 50% of the the lowest (i.e. optimal) delay in most cases. In some cases,
network capacity and manage the traffic to be with the algorithm finds the optimal solution very quickly;
minimum delay). The developed network could be unreliable however, there are some cases where longer convergence
in some sense because the reduced number of links may times are observed in order to obtain the optimal solution.
cause a problem if failure occurs in a link. This could be
another objective to investigate in the future. Of course the 6. Conclusions and Future Work
best reliable network will be the fully connected network. A Hierarchical Genetic Algorithms (HGA) based
This will be, of course, a very expensive network to optimization mechanism for Asynchronous Transfer Mode
implement. The computed delay times for various population (ATM) network has been formulated. Minimizing the total
sizes of the GA in levels 1 and 2 are shown in Table 3. cost is mainly the purpose of the proposed approach,
subsequently; the work addresses the maximum allowable
flow assignment in each link while simultaneously keeping
Table 3: Delay time for various Pop. sizes of GA levels the overall delay within the minimum acceptable range. It
can be inferred from the results obtained that ATM network
Description GAs level 1 GAs level 2 Delay time design using HGA produces good network plans in terms of
Pop. Size 10 25 10.81690 network throughput and GA utilization. Also, it allows more
Pop. Size 10 50 11.02230 flexibility in the traffic management, and less complexity of
Pop. Size 10 75 10.76282 the planning task. It is suggested to improve the current
work by adding new issues related to dynamic capacity
Pop. Size 30 25 10.80090
allocation within the network. Additionally, improvement
Pop. Size 30 50 10.75810
can be achieved if using Parallel Genetic Algorithms
Pop. Size 30 75 10.75850 (PGAs). This way the computation time can be reduced to a
Pop. Size 50 25 10.76630 reasonable computational effort.
Pop. Size 50 50 10.76430
Pop. Size 50 75 10.76006 References
[1] L. Painton and J. Campbell, “Genetic algorithms in
It was found that the minimum delay was achieved with optimization of system reliability,” IEEE Transactions
population size 30 for GA level 1 and 50 for GA level 2. In on Reliability, vol. 44, pp. 172–178, 1995.
general, as a common practice, the convergence of the GA
 (IJCNS) International Journal of Computer and Network Security, 61
Vol. 2, No. 3, March 2010

[2] D. Raychaudhuri and D. Wilson, “ATM–based networks: Genetic algorithm and tabu search
transport architecture for multiservices wireless approach,” International Journal of Computer Science,
personal communication networks,” IEEE Journal On vol. 1, no. 3, 2006.
Selected Areas In Communications, vol. 12, no. 8, pp. [18] R. Susmi, A. M. Sherry, and B. V. Reddy, “ATM
1401–1413, 1994. network planning: A genetic algorithm approach,”
[3] H. El-Madbouly, “Design and bandwidth allocation of Journal of Theoretical and Applied Information
embedded ATM networks using genetic algorithm,” In Technology, vol. 1, pp. 74–79, 2007.
Proceedings of World Academy of Science, [19] D. Thierens and D. E. Goldberg, “Convergence models
Engineering and Technology, vol. 8, pp. 249–252, of genetic algorithm selection schemes,” In PPSN III:
2005. Proceedings of the International Conference on
[4] D. W. Coit and A. E. Smith, “Reliability optimization Evolutionary Computation. The Third Conference on
of series-parallel systems using genetic algorithm,” Parallel Problem Solving from Nature. London, UK:
IEEE Transactions on Reliability, vol. 45, pp. 254–260, Springer-Verlag, pp. 119– 129, 1994.
1996. [20] B. L. Miller and D. E. Goldberg, “Genetic algorithms,
[5] Alaa Sheta, Mohammad Salamah and Malik Braik,” selection schemes, and the varying effects of noise,”
Topology Optimization and Traffic Management in Evolutionary Computation. vol. 4, no. 2, pp. 113–131,
Telecommunication Network Using Hierarchical 1996.
Genetic Algorithms”, In the Proceedings of
ICICIS2009, Ain Shams Univ., Cairo, Egypt, pp. 143- Authors Profile
148, March, 2009.
[6] E. lbaum and M. Sidi, “Topological design of local-
area networks using genetic algorithms,” In Alaa Sheta received his B.E., M.Sc.
IEEE/ACM Transactions on Networking, vol. 4, pp.
degrees in Electronics and Communication
766–778, 1996.
[7] K. Walkowiak, “Genetic algorithms for backup virtual Engineering from Faculty of Engineering,
path routing in survivable ATM networks,” In Cairo University in 1988 and 1994,
Proceedings of MOSIS 99, vol. 2, pp. 123–130, 1999.
respectively. A. Sheta received his Ph.D.
[8] A. Kasprzak and K. Walkowiak, “Algorithms for flow
assignment ATM virtual private networks,” In degree from the Computer Science
Proceedings Of First Polish-German Teletraffic Department, School of Information Technology and Engineering,
Symposium PGTS 2000, vol. 2, pp. 24–26, 2000.
[9] E. Alba, “Parallel evolutionary algorithms in George Mason University, Fairfax, VA, USA. His published work
telecommunications: Two case studies,” In Proceedings exceeds 70 publications between book chapters, journal papers,
of the CACIC02, Buenos Aires, Argentina, 2002. conferences and invited talks. His research interests include
[10] L. He and N. Mort, “Hybrid genetic algorithms for
telecommunications network back-up routing,” BT Evolutionary Computation, Computer Networks, Image Processing,
Technology Journal, vol. 18, no. 4, pp. 42–50, 2000. Software Reliability and Software Cost Estimation. Currently, Dr.
[11] L. Xian, “Network capacity allocation for traffic with
Sheta is a Professor with the College of Computers and Information
time priorities,” Journal of International Network
Management, vol. 13, pp. 411–417, 2003. Systems and the Director of the Vision and Robotics Laboratory at
[12] C. Rose and R. Yates, “Genetic algorithms and call Taif University, Taif, Saudi Arabia. He is on leave from the
admission to telecommunications networks,”
Electronics Research Institute (ERI), Giza, Egypt.
Computers and Operations Research, vol. 23, pp. 9–6,
1996.
[13] L. Davis and Coombs, Optimizing networks link sizes Malik Braik received his B.Sc. degree in
with genetic algorithms. Modeling and Simulation
Methodology, Knowledge System’s Paradigms, Electrical Engineering from Faculty of
Amsterdam, The Netherland: Elsevier, 1989. Engineering, Jordan University of Science
[14] S. Pierre and H. H. Hoang, “An artificial intelligence and Technology in 2000. Five years later,
approach for improving computer communication
network design,” Journal Of Operational Research M. Braik received his M.Sc in Computer
Society, vol. 41, no. 5, pp. 405–418, 1990. Science from Department of Information
[15] M. Gerla, J. A. S. Monteiro, and R. Pazos, “Topology
Technology, Al-Balqa Applied University, Jordan in 2005. His
design and bandwidth allocation in ATM nets,” IEEE.
JSAC, vol. 7, no. 8, pp. 1253–1262, 1989. research interests focus on Evolutionary Computation, Image
[16] A. Sheta, M. Salamah, H. Turabieh, and H. Heyasat, Processing, Data Security and Computer Networks. Currently,
“Selection of appropriate traffic of a computer network
Braik is working with the Department of Information Technology,
with fixed topology using GAs,” International Journal
of Computer Science, IAENG, vol. 34, no. 1, 2007. Prince Abdullah Bin Ghazi Faculty of Science and Information
[17] S. Routray, A. Sherry, and B. Reddy, “Bandwidth Technology, Al-Balqa Applied University, Al-Salt, Jordan.
optimization through dynamic routing in ATM
62 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

Mohammad Salameh received his B.E.

degree in Computer Science from Faculty of
King Abdullah II School for Information
Technology (KASIT), University Of Jordan
in 2005, and in 2009 Mohammad received
his M.Sc. degree in Computer Science from
Faculty of Graduate Studies and Scientific Research, Al-Balqa`
Applied University in 2009. Mohammad Salameh is a joiner
researcher; his research interest fields are Robotics Design and
Programming, Evolutionary Computation, Computer Networks,
Image Processing. Mohammad is a Robotics Trainer and
Programmer in the National Education Center for Robotics
(NECR), King Hussein Foundation, Amman, Jordan.
 (IJCNS) International Journal of Computer and Network Security, 63
Vol. 2, No. 3, March 2010

Capacity Enhancement of 3G Cellular System using

Switched Beam Smart Antenna with Windowed
Beam former
Prof. T. B. Lavate1, Prof. V. K. Kokate2 , Prof. Dr A. M. Sapkal 3
1
Department of E & T/C, College of Engineering Pune, Maharashtra, India,
[email protected], [email protected]
2
Department of E & T/C,Indira College of Engineering & Management, Pune, Maharashtra, India,
[email protected]
3
Department of E & T/C, College of Engineering Pune-5, Maharashtra, India,
[email protected]

Abstract: The 3G cellular networks are being designed to CDMA all subscribers use same frequency & this creates
provide high bit rate services, multimedia traffic, in addition to inter-cell & intra-cell co-channel interference. Thus the
voice calls within the limited bandwidth available. The suitable actual performance of CDMA system is still interference
solution to the bandwidth limitation is the smart antenna. The
limited and is affected by adverse channel conditions created
smart antennas at the base station of cellular system have the
interference rejection or signal to interference plus noise by multipath propagation. It is obvious that the capacity of
improvement capability and in turn to improve the capacity of CDMA based wireless communication system can be
cellular system. The smart antennas can generally be classified improved by different interference reduction techniques &
as either switched beam or adaptive array smart antenna. that can be achieved by using smart antennas such as
However switched beam smart antenna is cheaper to implement switched beam smart antenna (SBSA) or adaptive array
in many applications and hence investigated here much in smart antenna . In this paper section 2. describes switched
detail.
beam smart antenna used to improve the capacity of
Switched beam smart antenna (SBSA) creates a group of
CDMA cellular system.
overlapping beams that together result in omni directional
coverage. SBSA has the interference rejection capability
depending on side lobe level. The simplest way to reduce the side
lobe level and improve the SINR of SBSA is to use non adaptive
2. Switched Beam Smart Antenna and its
windowed beam forming functions. In this paper performance of Array Pattern
linear SBSA has been investigated using MATLAB for various
windowed beam forming functions such as Hamming, Gaussian 2.1 Concept of SBSA
and Kaiser-Bessel functions. It has been observed that the Depending upon the various aspects of smart antenna
Kaiser-Bessel weights provide one of which lowest array side
technology they are categorized as switched beam smart
lobe levels still maintaining nearly the same beam width as
uniform weights and consequently Kaiser-Bessel function can antenna and adaptive array smart antenna. Here we
widely be used with SBSA to improve the capacity of 3G cellular investigate switched beam smart antenna in detail. The
system. switched beam smart antenna (SBSA) has multiple fixed
beams in different directions & this can be accomplished
Keywords: SBSA, DS/CDMA, Windowed Beam forming using feed network referred to as beam former & most
commonly used beam former is Butler matrix. In terms of
1. Introduction radiation patterns switched beam is an extension of cellular
Nowadays the wireless operators are faced with increasing sectorization method of splitting a typical cell. That is
capacity demands for both voice & data services. To achieve switched beam antenna increases the capacity of cellular
this, multiple access techniques such as TDMA & CDMA system by creating micro sectors and each micro sector
can be employed. For instance in TDMA based wireless contains a pre-determined fixed beam pattern with greatest
communication system (2G systems) the frequency reuse sensitivity located in the centre of the beam & less sensitivity
factor is normally greater than 6. That is these systems elsewhere. The receiver selects the beam that provides
employ the frequency reuse concept which increases their greatest signal enhancement & interference reduction.
capacity to some extent. But this frequency reuse concept Actually SBSA enhances receive signals & switch from one
creates the co-channel interference. In CDMA systems the beam to another as the desired user moves throughout the
frequency reuse factor is 1 which enables it to offer higher sector. An N beam switched beam antenna generally
capacities & the capacity improvement by CDMA provides an N-fold antenna gain and some diversity gain by
technology may be as high as 13 times that of TDMA. But in combining the received signal from different beams as
64 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

shown in Fig.1. as shown in Fig. 2. It is obvious that N element SBSA forms

N spatial channels whose input signal is given by

Beam-1
S = [S1 S2 S3 ------- SN ]
(3)

Beam-2 B where Si = √Pi*[1 e-jφi e-j2φi e-j3φi ……. e-j (N-1)φi ] is the
e ith user signal φi=2πd/λ*sinΦ i ; Φi is the ith user angle of
a
m arrival signal with power Pi. Taking into account an
Signal output equation (1) and equation (3) the output of SBSA is
Interference f Beam
Signal o Select
r Y=AHS = [Y1 Y2 Y3 ----------YN]
m
Beam-3 e (4)
r

Beam-4 If all users are located exactly as өi = Φi; the vector Yi=[0, 0,
0, 0, √Pi, 0, 0, --- 0]. But if өi ≠ Φi; , we have to evaluate
Desired signal Direction the maximum element of each vector to detect active users.

0
Figure 1. Switched beam smart antenna.
-5

However these switched beam smart antennas have non -10

uniform gain with respect to AOA due to scalloping & they

Norm aliz ed power gain(dB)

-15
can also have a problem with locking into the wrong beam
-20
due to multi path fading or interference and provide limited
interference suppression since they can't suppress the -25

interference if it is in the same beam as the desired signal. -30

Hence switched beam smart antenna solutions work best in -35
minimal to moderate co-channel interference scenario.
-40
Further they are often much less complex and are easier to
retro-fit to existing wireless technologies. -45

-50
2.2 Signal model and array pattern of SBSA -60 -40 -20 0 20 40 60
AOA
The SBSA creates a number of two way spatial channels on
a single conventional channel in frequency, time or code. Figure 2. SBSA Array pattern for number of antenna
Each of these spatial channels has the interference rejection elements N = 8.
capabilities of the array depending on side lobe level (γ).The It is apparent from Fig.2 that the array factor of SBSA has
steering matrix for N element SBSA is given by side lobe level γ= -16 dB. The presence of side lobes means
A= [a1 a2 a3 ------- aN ] (1) the array radiates in untended directions or the side lobes
can receive the same signal from multiple angles which may
cause fading in communication systems. These harmful side
where ai = [1 e-jΨi e-j2Ψi e-j3Ψi ……. e-j (N-1)Ψi ] is the lobes of SBSA can be suppressed by windowing the array
steering vector; Ψi=2πd/λ*sinөi ; өi is the ith reference elements also called as array weighting and is discussed in
angle, d is the spacing between the antenna elements. The section 3.
matrix A forms the special filters with orthogonal properties
aiH ak =N, if i=k; and aiH ak =0, if i≠k. 3. WINDOWED BEAM FORMERS FOR SBSA
In practice the SBSA creates several simultaneous fixed
beams through the use of equation (1) and Butlar Matrix
y
theory. With Butlar Matrix for SBSA of N elements the
array factor can be given as
sin[Nπd/λ (sinө - sinөℓ)]
AF(ө) = (2)
Ѳ
Nπd/λ (sinө - sinөℓ) wN/2 w2 w1 w1 w2 wN/2
x
d
where sinөℓ = ℓλ/Nd; ℓ= ±1/2, ±3/2, ------- ± (N-1)/2. If
the element spacing is d = λ/2 the beams of SBSA are evenly
distributed over the span of 1800 and using MATLAB 7.0 Figure 3. N Element linear antenna array with
equation (2) for N=8 is simulated and simulation results are weights.
 (IJCNS) International Journal of Computer and Network Security, 65
Vol. 2, No. 3, March 2010

The array factor of N element uniform linear array is given

by
0
sin[NΨ/2]
Hamming window
AF(ө) = (5)
Boxcar window
N (sinΨ/2)
-10

where Ψ= kdsinө + β, β is the phase shift from element to

element. But such uniform linear array also called Boxcar -20
window, exhibits side lobe levels of about γ = -16 db. The
array side lobes can be suppressed by weighting or

|A F |d B
windowing the array elements as shown in Figure 3. The -30
array factor of such N linear element windowed array is
given by
-40

N/2
AFn (ө) = Σ wncos((2n-1/2)kdsinө)
-50
(6)
n=1
To determine the weights wn there are number of useful
-60
window functions that can provide weights for each element -90 -60 -30 0 30 60 90
in array viz. Hamming, Gaussian, Kaiser-Bessel functions θ
etc. By carefully controlling the side lobes in windowed
Figure 4. Array factor with Hamming weights for N=8
SBSA arrays, most interference can be reduced to
insignificant level. But side lobe suppression is achieved at
the expense of main lobe beam width. These windowed non 3.1.2 Gaussian weight function: The Gaussian
adaptive beam formers have advantages as: weights are found by Gaussian function to be

• Cheaper to implement than adaptive beam forming, w (n+1)= exp((-0.5(α (n-N/2)/N/2) 2)

• No main lobe distortions due to interference, (8)
• Any level of side lobe suppression is possible with correct where n=0,1,…,N α≥2
choice of windows.
The disadvantages of windowed beam former are: The Gaussian array weights can be determined using the
• Distribution and power of interference affects its gaussian(N) command in MATLAB. The normalized
performance, gaussian array weights for N=8 and α=2.5 are
• Increase in main lobe beam width (∆).
w1=1, w2=0.6766, w3=0.3098 and w4=0.0960.
3.1 Window weight functions for SBSA
The useful window functions for SBSA are investigated The Gaussian weighted array factor is plotted in Fig..5
below:
0
Gaussian window
3.1.1 Hamming weight function : The Hamming
Bxcar window
weights are chosen by -10

w(n+1)=0.54 -0.46 cos[2πn/(N-1)] (7)

-20
where n = 0, 1, 2, ……. ,N-1.
The hamming array weights can be found using the
|A F |d B

-30
hamming(N) command in MATLAB. Array pattern for N=8
elements with normalized Hamming weights for linear array
-40
is plotted in Fig. 4. It is apparent that Hamming weights
provide side lobe suppression γ = -39 db at the expense of
large increase in beam width ∆=2. -50

Table 1: Hamming normalized weights for array N=8

-60
-90 -60 -30 0 30 60 90
θ

Weight types w1 w2 w3 w4 Figure 5. Array factor with Gaussion weights for N=8
It is clear that Gaussian weighted array pattern provides side
lobe suppression γ = -48dB with increase in main lobe beam
Hamming 1 0.673 0.2653 0.0838
width ∆=1.85
66 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

3.1.3 Kaiser-Bessel weight function: The Kaiser lobe beam width (∆=1.2) compared to Hamming and
Bessel weights are determined by Gaussian weight functions & hence Kaiser Bessel weight
functions are chosen to suppress the side lobes in switched
beam smart antenna. (SBSA)
I0
w (n) =
(9) 4. Performance analysis of Kaiser-Bessel
I0
windowed SBSA and simulation results
We consider the DS/CDMA system model in which the data
where n= 0,…..,N/2, α>1 is modulated using BPSK format. The pulse and PN code
amplitudes are all independently and identically distributed
The Kaiser-Bessel normalized weights for N=8 are found random variables. We assume that the PN code length
using the Kaiser(N,α) command in MATLAB . M=128 & the power of each mobile station is perfectly
controlled by the same base station BS which employs
Table 2: Kaiser-Bessel normalized weights for array switched beam smart antenna.
N=8
As shown in [2] the bit error rate (BER) for DS-CDMA,
Weight types w1 w2 w3 w4 1200 sectorized systems is given by
Kaiser-Bessel 1 0.8136 0.5136 0.210

The Kaiser-Bessel weighted array factor is plotted in Fig 6 Pe = Q (10)

using MATLAB; in which γ = -33 db and ∆=1.2.
where Eb(1)/N0 is SINR for user of interest #1, Eb(k)/N0 is the
0
Kaiser-Bessel
same for interfering users.
Boxcar window We extended equation (10) to switched beam smart antenna
-10
as

-20
Pe= Q (11)
|A F |d B

-30

where k1 is the number of interfering users with same PN

-40 code like user #1 affected side lobe, k2 is the number of
interfering users affected main lobe & k3 is like k2 but
affected side lobes.
-50

0
10
-60 120 deg. sect. Antenna
-90 -60 -30 0 30 60 90 Boxcar SBSA
θ -2 Kaiser-Bessel SBSA
10
Figure 6. Array factor with Kaiser-Bessel weights N=8

-4
Table 3: Important parameters of windowed beams for 10
Bit Error Rate

N=8 element linear array

Window Beam width γ
-6
(∆) (db) 10
Boxcar 1.0 -16
Hamming 2.0 -38 -8
10

Gaussian 1.85 -48

(α = 2.5)
-10
Kaiser-Bessel 1.2 -33 10
0 2 4 6 8 10 12 14
(α = 3.0) Eb/No dB

Figure 7. The BER performance for DS/CDMA and

Table3. shows that Kaiser-Bessel function provides side lobe switched beam smart antenna for 100 users and N=8
suppression γ = -33dB but with minimum increase in main
 (IJCNS) International Journal of Computer and Network Security, 67
Vol. 2, No. 3, March 2010

For Boxcar (non windowed) SBSA the side lobe level (as [5] Md. Bhakar, Vani R. M., P. V. Hunagund, “Smart
shown in fig.2) γ = -16 db. While as Kaiser-Bessel weights Antennas Systems: Concepts of Beam steered Array
shown in Table 2 can be selected for SBSA so that its side Configurations”, Proceedings of IEEE International
lobe level (as shown in fig.6) can be reduced to γ= -33db. symposium on microwaves 2008, Banglore section
Using MATLAB 7.0 equation (10) and equation (11) with [6] David Cabrera, Joel Rodriuez,Adviser:Victor
γ= -16 db. & γ= -33db are simulated. V.Zaharov, “ Switched Beam Smart Antenna BER
Performance analysis for 3G CDMA Cellular
That is using equation (10) & (11) the simulation results are Communication”, Polytechnic University of Puerto
presented in Fig 7. where BER as a function of Eb/N0 and Rico.
number of active users = 100 in the service area is plotted.
Fig. 7 presents BER of DS/CDMA system with conventional Authors Profile
1200 sectored antenna, no windowed SBSA (N=8, γ= -16dB)
& Kaiser Bessel windowed SBSA (N=8, γ = -33dB). As
follows from Fig. 7 for the fixing level of BER, e.g. for 3G T.B. Lavate received ME (Microwave)
communication system acceptable BER is Pe =10-4 , by degree in E&T/C in 2002 from Pune
the application of SBSA the number of active users in University. He is pursuing his Ph.D. in
conventional CDMA system can be increased up to 1.12 College of Engineering, Pune-5 affiliated to
times for no windowed SBSA antenna and up to 1.4 times pune university. He has published nine
with Kaiser- Bessel windowed SBSA antenna. i.e Kaiser- papers on wireless communication & smart
Bessel windowed SBSA antenna can increase the number of antenna. He is member of IETE & ISTE.
active users in 3G system significantly without losing of
performance quality.
V. K. Kokate graduated in E & T/C from
University of Pune in 1973 and received his
5. Conclusion Masters in E & T/C (Spl: Microwave) from
In this paper the performance analysis of windowed SBSA is the University of Pune..
investigated. The relation of BER for DS/CDMA is extended Presently he is working as H.O.D. of E&TC Engineering
for the case of application of SBSA at base station. As a Department of Indira College of Engineering, , Pune. having
over 35 years experience in teaching and administrative. His
result shows, the application of Kaiser-Bessel windowed
fields of interests are Radar, Microwaves, Antennas and
SBSA improves the BER which in turn increases the active
EMI & C. He is a Fellow Member of IETE and Member
users in conventional CDMA system significantly & hence
of ISTE/IEEE.
SBSA is still an attractive solution to increase the capacity of In his credit, he has about twenty papers published in
existing 3G cellular wireless communication system. International/National repute Conferences/ Journals.

References A.M.Sapkal graduated in E&TC from

University of Pune in 1987 and received
[1] L.C.Godara,“Application of antenna arrays to mobile his Masters in E & TC (Spl: Microwave)
communications II: Beamforming & direction of from the University of Pune in 1992 and
arrival conciderations”, Preceedings of IEEE,volume obtained his Ph.D in 2008. Presently he is
85,issue 8,August 1997, pages 1195-1245. working as Professor in Dept. of E &TC Engineering of
[2] T. S. Rappaport,“ Smart antennas for wireless College of Engineering, Pune. Having, over 20 years
communications,” Prentice Hall, India 2005. experience in teaching, his fields of interests are power
[3] Frank Gross ,“ Smart antennas for wireless electronics, microwaves, and image processing. He is a
communication,” McGraw-Hill, New York 2005 member of ISTE, IETE & IEEE.
[4] J. C. Liberti, T. S. Rappaport, “Smart antennas for In his credit, he has about twenty five papers published in
wireless communications: IS-95 and Third Generation National / International repute Conferences/Journals.
CDMA Applicatons”, Prentice Hall PTR, New Jersey
1999..
68 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

DDOM: The Dynamic Data Oriented Model for

Image Using Binary Tree
A.Shahrzad Khashandarag1*, A.Mousavi2, R.Aliabadian3, D. Kheirandish4 and A.Ranjide Rezai5
1
Young Researchers Club of Tabriz, Islamic Azad University Tabriz Branch,
Tabriz-Iran
1
[email protected]
2 3 4 5
[email protected] [email protected] [email protected] [email protected]

GIF (Graphics Interchange Format) is an 8-bit-per-pixel

Abstract: This paper presents a dynamic data oriented
model (DDOM) of image. The ability of model is very fast image bitmap image format that was introduced by CompuServe in
processing against Data oriented Model that Habibizad and 1987 and has since come into widespread usage on the
cooperators proposed in [8]. In our approach, the Sobel World Wide Web due to its wide support and portability.
algorithm is used for edge detection of image and histogram
thresholding is used for clustering of image and binary tree is The format uses a palette of up to 256 distinct colors from
used for anatomy of model. Each node of this tree represents the the 24-bit RGB color space. GIF images are compressed
feature of image or sub image. By using the presented model in using the LZW lossless data compression technique to
this paper, tasks as measuring similarity ratio, segmentation, reduce the file size without degrading the visual quality
clustering … can be done with high desired precision and [10].
corresponding speed [8].
Keywords: Data oriented modeling; Image Segmentation; In computing, JPEG (pronounced JAY-peg) is a commonly
histogram thresholding; Binary Tree. used standard method of compression for photographic
images. The name JPEG stands for Joint Photographic
1. Introduction Experts Group, the name of the committee who created the
Recently many researchers have been studied on data standard. The group was organized in 1986, issuing a
structures for image processing tasks. In [1-7] has been standard in 1992, which was approved in 1994 as ISO
explained the data structures of images. The main idea of 10918-1. The compression method is usually lossy
this paper is to present an optimized and dynamic data compression, meaning that some visual quality is lost in the
oriented model of image. Dynamic Data Oriented Modeling process, although there are variations on the standard base
(DDOM) is an approach which models concepts by using line JPEG, which are lossless. There is also an interlaced
the data structure. We have introduced dynamic data "progressive" format, in which data is compressed in
oriented modeling of image for fast image processing. The multiple passes of progressively higher detail. This is ideal
Conventional methods use the color of the pixels to model for large images that will be displayed whilst downloading
the image and try to reduce the size of image (Compression) over a slow connection, allowing a reasonable preview
[8]. Previous image models are: before all the data has been retrieved [11].

• BMP DOM (Data Oriented Model) proposed by Habibizad and

• JPEG cooperators. This model is design for tasks that are very fast.
In the related works section, this model will be explained.
• GIF
• DOM
2. Related Works
.BMP or .DIB (device-independent bitmap) is a bitmapped In this paper we want to optimize the model of DOM which
graphics format used internally by the Microsoft Windows the Habibizad and cooperators proposed in [8]. Habibizad
and OS/2 graphics subsystem (GDI), and used commonly as and cooperators in [8] proposed:
a simple graphics file format on those platforms. This
format is in two parts, Part one, is the data, which indicate
whole image features including width, height, color palette 2.1 Data Oriented Modeling
etc. The second part consists a block of bytes that describes Habibizad and cooperators suppose that Fig1 is the image
the image, pixel by pixel. Pixels are stored starting in the for data oriented modeling. Fig. 2 shows ADBT of Fig. 1.
bottom left corner going from left to right and then row by ADBT is Average-Difference-Binary-Tree-of-Image which
row from the bottom to the top. Each pixel is described is stored in an array. This ADBT has three levels, numbered
using one or more bytes [9]. as 0, 1 and 2. The features of pixels are stored in the leaves.
Features of F are obtained by combining features of A and
C. In the same way, features of G are obtained from B and
 (IJCNS) International Journal of Computer and Network Security, 69
Vol. 2, No. 3, March 2010

D. finally features of the entire image, E, are obtained by Fig. 6 shows segmented of Fig. 5. This segmented image is
combining features of F and G. By putting F and G together, created with histogram thresholding. Fig. 7 shows the
we can achieve smooth of original image as is shown in Fig. reverse N for image processing in this model. Fig. 8 shows
3 [8]. ADBT of Fig. 6. This ADBT has three levels, numbered as
0, 1 and 2. The features of regions are stored in the leaves.
Fig. 4 shows the array, which stores the ADBT of Fig. 2. Features of L0 are obtained by combining features of R0 and
R1. In the same way, features of L1 are obtained from R2
and R3. Finally features of the entire image, L2, are
obtained by combining features of L0 and L1. By putting L0
and L1 together, we can achieve smooth of original image
as is shown in Fig. 9. Fig. 10 shows the ADBT of Fig. 6 that
1
Figure 1. an I 2 Image [8]. has been stored in an array.

In Fig.8 A and B and D have the similar intensity.

3.2 Edge detection with Sobel Algorithm

Edges characterize boundaries and are therefore a problem
of fundamental importance in image processing. Edges in
imag-

1
Figure 2. ADBT of an I2 image [8].

Figure 3. Smoothed of original image [8].

Figure 6. a segmented image.

Figure 4. ADBT of Fig. 2 stored in an array [8].

For ease of understanding, mentioned steps are described

top to down; but for enhancing speed of making ADBT, it
will be created down to top [8]. After that leaves will be
initialized by values of produced vector. Therefore, count of
leaves is equal to count of pixels of image. Each leaf is
corresponded to a pixel of original image and its A and D
are same as the pixel’s color. For each non-leaf node, A is
equal to average of its children’s A and D is equal to
difference of its children’s A [8].
Figure 7. the reverse N for image processing.

3. Our Works

3.1 Dynamic Data Oriented Modeling

This paper presents a dynamic data oriented model
(DDOM) of image. To illustrate the concept, Fig. 5 shows a
3 3 images, which have nine pixels, labeled A, B, C, D, E, F,
G, H and I respectively.

Figure 8. ADBT of Fig.8

Figure 9. Smoothed of original image

Figure 5. An instance Image
70 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

Figure 10. ADBT of Fig. 8 stored in an array

-es are areas with strong intensity contrasts – a jump in

intensity from one pixel to the next. Edge detecting an
image significantly reduces the amount of data and filters
out useless information, while preserving the important
structural properties in an image. There are many ways to
perform edge detection. However, the majority of different
methods may be grouped into two categories, gradient and
Laplacian. The gradient method detects the edges by looking
for the maximum and minimum in the first derivative of the
image. The Laplacian method searches for zero crossings in
the second derivative of the image to find edges [12].

The Sobel operator performs a 2-D spatial gradient

measurement on an image. The Sobel operator is shown in
Fig. 11.
Figure 12. Edge detection with Sobel algorithm
Note: G is Gradient and f is Image.
4.3 Image Segmentation Algorithm
According to the S. Arora and cooperator’s results in article
[13], segmentation algorithm is:

Following steps describe the proposed algorithm for image

segmentation:

1. Repeat step 2–6, times; where n is the

Figure 11. Sobel Operator. number of thresholds.
2. Range R = [a, b]; initially a = 0 and b = 255.
(1) 3. Find mean and standard deviation of all the
(2) pixels in R.
4. Sub-ranges’ boundaries and are calculated as
and ;
where and are free parameters.
5. Pixels with intensity values in the interval [ ]
(3)
and [ ] are assigned threshold values equal to
(4) the respective weighted means of their values.
6. a = T1 + 1 ; b = T2 − 1 .
7. Finally, repeat step 5 with and
with .
(5)
In Fig13 shown the result of image segmentation algorithm.
(6) Also in Fig14 the result of histogram Lena is shown.

In Fig. 12 Shown the result of Sobel algorithm:

4. Conclusion
In this paper, a dynamic data oriented model for image is
introduced which models image as a binary tree. Each node
of this tree represents the feature of image or sub image. By
using the presented model in this paper, tasks as measuring
similarity ratio, segmentation, clustering … can be done
with high desired precision and corresponding speed.
 (IJCNS) International Journal of Computer and Network Security, 71
Vol. 2, No. 3, March 2010

segmentation through a fast statistical recursive

algorithm", pattern recognition letters 29 (2008) 119-
125, published by Elsevier.

Authors Profile
Asghar Shahrzad Khashandarag received the
B.S. degree in computer engineering from the
Payame Noor University Bonab Branch, Iran, in
2008, and the M.S. degree in computer
engineering from the Islamic Azad University
Tabriz Branch, Iran, in 2009, respectively.
From 2008, he works as a researcher with the Young Researchers
Club of Tabriz. He has published more than 10 papers in various
journals and conference proceeding. His research interests include
Figure 13. Results: Lena (a) Lena gray, (b) histogram, (c) 2 image processing, signal processing, wireless sensor network.
level thresholding, (d) 4 level, (e) 6 level and (f) 8 level.
Alireza Mousavi received his B.Sc. in computer
engineering, software Engineering, from
Allameh Mohaddes Nouri University,
Mazandaran, Iran, in 2008, and the M.S. degree
in computer engineering from the Islamic Azad
University Tabriz Branch, Iran, in 2010,
respectively. His research interests include image processing,
Residue Number systems, wireless sensor network.

Figure 14. Result of Histogram Ramin Aliabadian received the B.Sc. degree in
computer engineering from the Shomal
University, Amole, Iran, in 2008, and the M.S.
degree in computer engineering from the Islamic
Azad University Arak Branch, Iran, in 2010,
References respectively. His research interests include
[1] Zhiyong Wang, Dagan Feng, and Zheru Chi, “Region- image processing, computer architecture, computer networks.
Based Binary Tree Representation For Image
Classification”, IEEE International Conference on
Neural Networks & Signal Processing, Nanjing,
China, 2003.
[2] Xiaolin Wu, “Image Coding by Adaptive Tree-
Structured Segmentation”, IEEE Transactions on Davar Kheirandish Taleshmekaeil received
Information Theory, VOL. 38, NO. 6, 1992. the B.Sc. degree in Computer Hardware
[3] G.S.Seetharaman, B.Zavidovique, “Image Processing in engineering from the Allameh Mohaddes Nouri
a Tree of Peano Coded Images”, Computer Architecture University, Mazandaran, Iran, in 2008, and the
for Machine Perception, 1997.CAMP’97. Proceeding M.S. degree in computer engineering from
Fourth IEEE International Workshop on.
[4] M. Kunt, M. Benard, and R. Leonardi, “Recent results the Islamic Azad University Tabriz Branch,
in highcompression image coding”, IEEE Transaction Tabriz, Iran, in 2010, respectively. His research interests include
Circuits Syst., vol. CAS- 34, no. 11, pp. 1306-1336, image processing, computer architecture, computer networks.
Nov 1987.
[5] R. Leonardi and M Kunt, “Adaptive split-and-merge for
image analysis and coding”, Proc. SPIE, vol. 594, 1985.
[6] G. J. Sullivan and R. L. Baker, “Efficient quadtree
coding of images and video”, in ICASSP Proc., May Ali Ranjide Rezai received the B.Sc. degree in
1991, pp. 2661-2664. computer engineering from the Shomal
[7] J. Vaisey and A. Gersho, “Image compression with University Amole, Iran, in 2008, and the M.S.
variable block size segmentation”, IEEE Trans. Signal
Processing, vol. SP-40, pp.2040- 2060, Aug. 1992. degree in computer engineering from the
[8] A. Habibizad Navin, A. Sadighi, M. Naghian Islamic Azad University Tabriz Branch, Iran, in
Fesharaki, M. Mirnia, M. Teshnelab, and R. Keshmiri, 2010, respectively. His research interests
"Data Oriented Model of image : as a framework for include image processing, computer architecture.
image processing", World Academy of Science,
Engineering and Technology 34 2007
[9] http://en.wikipedia.org/wiki/Windows_bitmap
[10] http://en.wikipedia.org/wiki/GIF
[11] http://en.wikipedia.org/wiki/JPEG
[12] http://www.pages.drexel.edu/~weg22/edge.html
[13] S. Arora, J. Acharya , A. Verma, Prasanta K.
Panigrahi, " Multilevel thresholding for image
72 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

Comprehensive Analysis and Enhancement of

Steganographic Strategies for Multimedia Data
Hiding and Authentication
Ali Javed, Asim Shahzad, Romana shahzadi, Fahad Khan

Faculty of Telecommunication and Information Engineering

University of Engineering and Technology Taxila Pakistan
[email protected], [email protected], [email protected]

Abstract: This research paper focuses on the analysis and wish to plan an escape from jail.
enhancement of steganographic strategies for multimedia data
hiding authentication. Based on an authentication game
between an image and its authorized receiver, and an opponent,
security of authentication watermarking is measured by the
opponent's inability to launch a successful attack. In this work,
we consider two stages of data hiding mechanism: Hiding the
data in an image along with conditional security and detecting
the hidden data. First we detect whether there exists a hidden
message within the image and then applying the conditional
security mechanism, we extract that hidden message. We
propose a novel security enhancement strategy that results in
efficient and secure LSB-based embedding and verification
phenomenon. Both theoretical analysis and experimental results
are presented. They show that using our approach, protection is Figure 1. Stenography Process
achieved without significant increase in image size and color
distortion, and without sacrificing the image or video quality. However, the prison warden, Ward, can monitor any
communication between Alice and Bob, and if he detects any
Keywords: Steganography, LSB, Stego image, Stego key, hint of “unusual" communications, he throws them both in
payload, Watermarking, Covert Communication solitary confinement. Alice and Bob must then transmit their
secret plans so that nothing in their communication seems
1. Introduction unusual" to Ward. There have been many proposed solutions
to this problem, ranging from rudimentary schemes using
The word steganography literally means covered writing as invisible ink to a protocol which is provably secure assuming
derived from Greek. It includes a vast array of methods of that one-way functions exist.
secret communications that conceal the very existence of the
message. Among these methods are invisible inks, microdots,
character arrangement (other than the cryptographic methods
of permutation and substitution), digital signatures, covert
Channels and spread-spectrum communications.
Steganography is the art of concealing the existence of
information within seemingly inoffensive carriers.
Steganography can be viewed as parallel to cryptography.
Both have been used throughout recorded history as means to
protect information. At times these two technologies seem to
converge while the objectives of the two differ.
Cryptographic techniques "scramble" messages so if
intercepted, the messages cannot be understood.
Steganography, in an essence, "camouflages" a message to
hide its existence and make it seem "invisible" thus
concealing the fact that a message is being sent altogether. Figure 2. General Model of Steganography
An encrypted message may draw suspicion while invisible
messages will not [1]. Steganography refers to the problem of 2. Literature Survey
sending messages hidden in “innocent looking"
communications over a public channel so that an adversary 2.1 Steganography Techniques
eavesdropping on the channel cannot even detect the presence
2.1.1 Physical steganography
of the hidden messages. Simmons gave the most popular
Steganography has been widely used including recent
formulation of the problem: two prisoners, Alice and Bob[2],
historical times and the present day. Possible permutations
 (IJCNS) International Journal of Computer and Network Security, 73
Vol. 2, No. 3, March 2010

are endless and known examples include • Chaffing and winnowing.

•Hidden messages within wax tablets: in ancient Greece, • Mimic functions convert one file to have the statistical
people wrote messages on the wood, and then covered it with profile of another. This can thwart statistical methods that
wax upon which an innocent covering message was written. help brute-force attacks identify the right solution in a
• Hidden messages on messenger's body: also in ancient ciphertext-only attack.
Greece. Herodotus tells the story of a message tattooed on a • Concealed messages in tampered executable files,
slave's shaved head, hidden by the growth of his hair, and exploiting redundancy in the i386 instruction set.
exposed by shaving his head again. The message allegedly • Pictures embedded in video material (optionally played at
carried a warning to Greece about Persian invasion plans. slower or faster speed).
This method has obvious drawbacks such as delayed • Injecting imperceptible delays to packets sent over the
transmission while waiting for the slave's hair to grow, and network from the keyboard. Delays in key presses in some
its one-off use since additional messages requires additional applications (telnet or remote desktop software) can mean a
slaves. In WWII, the French Resistance sent some messages delay in packets, and the delays in the packets can be used to
written on the backs of couriers using invisible ink. encode data.
• Hidden messages on paper written in secret inks, under •Content-Aware Steganography hides information in the
other messages or on the blank parts of other messages. semantics a human user assigns to a datagram. These systems
• Messages written in morse code on knitting yarn and then offer security against a non-human adversary/warden.
knitted into a piece of clothing worn by a courier. •Blog-Steganography. Messages are fractionalized and the
• Messages written on the back of postage stamps. (encrypted) pieces are added as comments of orphaned web-
logs (or pin boards on social network platforms). In this case
2.1.2 Digital steganography
the selection of blogs is the symmetric key that sender and
Modern steganography entered the world in 1985 with the advent
recipient are using; the carrier of the hidden message is the
of the personal computer applied to classical steganography
problems. Development following that was slow, but has since
whole blogosphere.
taken off, going by the number of 'stego' programs available: Over
2.2 Types of Steganography
725 digital steganography applications have been identified by the
Steganography Analysis and Research Center. Pure Steganography
Pure steganography is where you only need to know the
technique to be able to read off the message.
Private Steganography
Private Steganography requires you to know a password as
well as the technique.
Public steganography
Public steganography is like public key cryptography - you
have a public and private key and knowledge of the
technique. A standard library for pure steganography isn't so
smart - because it would allow people to simply try the
techniques.

2.3 Image Steganography

Figure 3. Cover Image
Steganography is the art of hiding the fact that
Here a cover Image of a tree is shown in fig. 3. By removing
communication is taking place, by hiding information in
all but the last 2 bits of each color component, an almost
other information. Many different carrier file formats can be
completely black image results. Making the resulting image
used, but digital images are the most popular because of their
85 times brighter results in the image below.
frequency on the Internet.
For hiding secret information in images, there exist a large
variety of steganographic techniques some are more complex
than others and all of them have respective strong and weak
points.
As stated earlier, images are the most popular cover objects
used for steganography. In the domain of digital images many
different image file formats exist, most of them for specific
applications. For these different image file formats, different
steganographic algorithms exist.

2.4 Image Compression

Figure 4. Stego Image When working with larger images of greater bit depth, the
images tend to become too large to transmit over a standard
Digital steganography techniques include: Internet connection. In order to display an image in a
• Concealing messages within the lowest bits of noisy images reasonable amount of time, techniques must be incorporated
or sound files. to reduce the image’s file size. These techniques make use of
• Concealing data within encrypted data. The data to be mathematical formulas to analyze and condense image data,
concealed is first encrypted before being used to overwrite resulting in smaller file sizes. This process is called
part of a much larger block of encrypted data.
74 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

compression. In images there are two types of compression,

Lossy and Lossless. Both methods save storage space, but the
procedures that they implement differ. Lossy compression
creates smaller files by discarding excess image data from the
original image. It removes details that are too small for the
human eye to differentiate, resulting in close approximations
of the original image, although not an exact duplicate. An
example of an image format that uses this compression
technique is JPEG (Joint Photographic Experts Group).
Lossless compression, on the other hand, never removes any
information from the original image, but instead represents
data in mathematical formulas. The original image’s integrity
is maintained and the decompressed image output is bit-by-
bit identical to the original image input. The most popular
image formats that use lossless compression is GIF
(Graphical Interchange Format) and 8-bit BMP (a Microsoft Figure 5. Main Flow Diagram
Windows bitmap file).
3.2 Text in Image Flow Diagram
2.5 Video Steganography
Since a video can be viewed as a sequence of still images,
video steganography can be viewed simply as an extension of
image steganography. The internet and the World Wide Web
have revolutionaries the way in which digital data is
distributed. The widespread and easy access to multimedia
content has motivated development of technologies for digital
steganography or data hiding, with emphasis on access
control, authentication, and copyright protection.
Steganography deals with information hiding, as opposed to
encryption. Much of the recent work in data hiding is about
copyright protection of multimedia data. This is also referred
to as digital watermarking. Digital watermarking for
copyright protection typically requires very few bits, of the
order of 1% or less of the host data size. These watermarks
could be alpha-numeric characters, or could be multimedia
data as well.
One of the main objectives of this watermarking is to be able
to identify the rightful owners by authenticating the
watermarks. As such, it is desirable that the methods of
embedding and extracting digital watermarks are resistant to
typical signal processing operations, such as compression,
and intentional attacks to remove the watermarks.
While transparent or visible watermarks are acceptable in
many cases, hidden data for control or secure communication
need to be perceptually invisible. The signature message data
is the data that we would like to embed or conceal. The
source data is used to hide the signature data; we often refer
to the source as the host data. After embedding a signature in
to a host, we get the watermarked or embedded data. The
recovered data, also referred to as the reconstructed data, is
the signature that is extracted from the embedded data. [11]
Figure 7. Text in Image flow Diagram
3. Proposed Methodology 3.3 Image in Image Flow diagram
3.1 Proposed Main Flow Diagram
The proposed methodology in this work is to embed the stego
message into image and video using LSB Technique.
 (IJCNS) International Journal of Computer and Network Security, 75
Vol. 2, No. 3, March 2010

3.5 LSB Technique

The most widely used technique to hide data is the usage of
the LSB- Least Significant Bit technique. Least Significant
Bit insertion method is a simple approach to embed
information in a cover file [11]. The LSB is the lowest order
bit in a binary value. This is an important concept in
computer data storage and programming that applies to the
order in which data are organized, stored or transmitted [12].
Usually, three bits from each pixel can be stored to hide an
image in the LSBs of each byte of a 24-bit image.
Consequently, LSB requires that only half of the bits in an
image be changed when data can be hidden in least and
second least significant bits and yet the resulting stego-image
which will be displayed is indistinguishable to the cover
image to the human visual system [11].
When using a 24-bit image, a bit of each of the red, green
and blue color components can be used, since they are each
represented by a byte. In other words, one can store 3 bits in
each pixel.
An 800 × 600 pixel image, can thus store a total amount of
1,440,000 bits or 180,000 bytes of embedded data [13]. For
example a grid for 3 pixels of a 24-bit image can be as
follows:
(00101101 00011100 11011100)
Figure 7. Image in Image flow Diagram (10100110 11000100 00001100)
(11010010 10101101 01100011)
3.4 Text in Video Flow Diagram When the number 200, which binary representation is
11001000, is embedded into the least significant bits of this
part of the image, the resulting grid is as follows:

(00101101 00011101 11011100)

(10100110 11000101 00001100)
(11010010 10101100 01100011)

Although the number was embedded into the first 8 bytes of

the grid, only the 3 underlined bits needed to be changed
according to the embedded message. On average, only half of
the bits in an image will need to be modified to hide a secret
message using the maximum cover size [13]. Since there are
256 possible intensities of each primary color, changing the
LSB of a pixel results in small changes in the intensity of the
colors. These changes cannot be perceived by the human eye -
thus the message is successfully hidden. With a well-chosen
image, one can even hide the message in the least as well as
second to least significant bit and still not see the difference.
In the above example, consecutive bytes of the image data –
from the first byte to the end of the message – are used to
embed the information. This approach is very easy to detect
[14]. A slightly more secure system is for the sender and
receiver to share a secret key that specifies only certain pixels
to be changed. Should an adversary suspect that LSB
steganography has been used, he has no way of knowing
which pixels to target without the secret key [15].
In its simplest form, LSB makes use of BMP images, since
they use lossless compression. Unfortunately to be able to
hide a secret message inside a BMP file, one would require a
very large cover image.

Figure 8. Text in Video Flow Diagram 3.6 Implementation of LSB Technique

To illustrate implementation of LSB technique, consider the
following image of parrots showing true colors Palette of the
76 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

Image resulting difference between the new from the original color
value is called the embedding error. Since there are only
three LSB’s for each pixel, the total number of bits that can
be hidden is only three times the total number of pixels
having the dimensions 768x512.

4. Simulation and Results

Following images were taken and processed by the
application of Digital Steganography and the results are as
following:

4.1 Procedure for Text in Image hiding

Figure 9. Parrot (True Color Palette Image) The procedure is started with opening the image in which
This image is composed of red, green, and blue color you want to hide the data and then enter the stego message in
channels. The pixel at the top-left corner of the picture has the specified space. The image is shown in fig.11and the
the values 122, 119, and 92 for its red, green, and blue color stego message is shown in the text box.
components respectively. In binary, these values may be
written as:

01111010 01110111 01011100

Figure 11. Open Image

The stego message is then merged into the original image as

shown in fig. 12.

Figure 10. Red, Green and Blue color channels of Parrot

Image

To hide the character “a” in the image, the LSB (the

rightmost bit) of each of the three 8-bit color values above
will be replaced with the bits that form the binary equivalent
of the character “a” (i.e., 01100001). This replacement
operation is generally called embedding. After embedding,
the color value would now change to:
01111010 1110111 01011101
Since there are only three values, only three of the eight bits Figure 12. Merge Text
of the character “a” can fit on this pixel. Therefore the The stego message is detected on the receiving end as shown
succeeding pixels of this image will also be used. In the three in fig. 13.
color values shown above, only the last value actually
changed as a result of LSB encoding, which means almost
nothing has changed in the appearance of the image.
Nevertheless, even in case wherein all LSB’s are changed;
most images would still retain their original appearance
because of the fact that the LSB’s represent a very minute
portion (roughly 1/255 or 0.39%) of the whole image. The
 (IJCNS) International Journal of Computer and Network Security, 77
Vol. 2, No. 3, March 2010

mountain on the left is the original image in which the stego

image will be embedded.

Figure 13. Detect

Figure 16. Open Images
The stego message is finally extracted as shown in the text
box in fig. 14 For merging the stego image in the original image user has to
enter the password and then the stego image embeds into the
original image.

Figure 14. Extract

The difference between the original image and the image Figure 17. Merge Image
after insertion of stego message is shown in fig. 15. The two
images look exactly the same which is why image On the receiving end the stego image can be recovered by
steganography is so important and useful in hiding data or entering the password as shown in fig. 18.
sending secret messages

Figure 18. Enter Password

Figure 15. View Difference The stego image is finally extracted after entering the
password as shown in fig. 19

4.2 Procedure for Image in Image stenography

In this phase the stego message is also image. The image of a
cat in fig. 16 is a stego image in this case and the image of a
78 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

Figure 19. Extract Image

The difference between the original image and the image Figure 22. View Frames
after insertion of stego message is shown in fig. 20. The two
images look exactly the same. The stego text is added to be embedded in the video as shown
in fig. 23

Figure 20. View Difference

4.3 Procedure for Text in Video stenography

Figure 23. Add Text
In this phase the stego message is embedded in the video, as
you can see in fig. 21 the video is added to enter the stego The stego text is detected as shown in fig. 24
message.

Figure 24. Detect text

Figure 21. Add Video
The frames can be viewed in the fig. 22 Finally the stego data is extracted on the receiving end as
shown in fig. 25
 (IJCNS) International Journal of Computer and Network Security, 79
Vol. 2, No. 3, March 2010

[11] N. F. Johnson, S. Jajodia, “Exploring Steganography: Seeing

the Unseen,” IEEE Computer, February 1998, pp.26-34.
[12] Julie K. Petersen, The Telecommunications Illustrated
Dictionary, CRC Press, 2002, ISBN: 084931173X.
[13] Krenn, R., “Steganography and Steganalysis”.
[14] Wang, H & Wang, S, “Cyber warfare: Steganography vs.
Steganalysis”, Communications of the ACM, 47:10, October
2004
[15] Anderson, R.J. & Petitcolas, F.A.P., “On the limits of
steganography”, IEEE Journal of selected Areas in
Communications, May 1998
[16] E. Kawaguchi and R. O. Eason :"Principle and applications of
BPCS-Steganography", Proceedings of SPIE: Multimedia
Systems and Applications, Vol.3528, pp.464-463, 1998.

Figure 25. Extract

Author Profile
5. Conclusion
Engr. Ali Javed is serving as a Lecturer in
Thus we conclude that steganographic techniques can be Software Engineering Department at
used for a number of purposes along with the covert University of Engineering & Technology
communication or deniable data storage, Confidential Taxila, Pakistan since September, 2007.
communication, Protection of data alteration, Secret data He has received his MS degree in Computer
storing, Media Database systems, Copyright Protection, Engineering from the University of
Feature Tagging and perhaps most importantly digital Engineering & Technology Taxila, Pakistan
watermarking. A digital watermark is invisible to the eye, in January, 2010. He has received B.Sc. degree in Software
undetectable without the appropriate secret key, but contains Engineering from University of Engineering & Technology Taxila,
Pakistan, in September, 2007. His areas of interest are Digital
small ownership identification. A much wider field of
Image Processing, Computer vision, Video Summarization,
steganography is digital Steganography allows copyright
Machine Learning, Software Design and Software testing.
owners to incorporate into their work identifying information
invisible to human eye, yet protecting against the dangers of
copyright infringement.

References

[1] Neil F. Johnson, Zoran Duric, Sushil Jajodia, Information

Hiding: Steganography and Watermarking - Attacks and
Countermeasures Kluwer Academic Press, Norwrll, MA,
New York, The Hague, London, 2000.
[2] G. Simmons, \The prisoners problem and the subliminal
channel," CRYPTO, pp. 51{67, 1983.
[3] C. Kurak, J. McHugh, "A Cautionary Note On Image
Downgrading," IEEE Eighth Annual Computer Security
Applications Conference, 1992. pp. 153-159.
[4] Tuomas Aura, "Invisible Communication," EET 1995,.
[5] Warren Zevon, Lawyers, Guns, and Money. Music track
released in the albums Excitable Boy, 1978; Stand in the Fire,
1981; A Quiet Normal Life, 1986; Learning to Flinch, 1993.
[6] David Kahn, The Codebreakers, The Macmillan Company.
New York, NY 1967. [Kurak92] C. Kurak, J. McHugh, "A
Cautionary Note On Image Downgrading," IEEE Eighth
Annual Computer Security Applications Conference, 1992.
pp. 153-159.
[7] Herbert S. Zim, Codes and Secret Writing, William Marrow
and Company. New York, NY, 1948.
[8] Anderson, R.J. & Petitcolas, F.A.P., “On the limits of
steganography”, IEEE Journal of selected Areasin
Communications, May 1998
[9] Owens, M., “A discussion of covert channels and
steganography”, SANS Institute, 2002
[10] Johnson, N.F. & Jajodia, S., “Steganalysis of Images Created
Using Current Steganography Software”, Proceedings of the
2nd Information Hiding Workshop, April 1998
80 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

Modeling and Simulation of Synchronous Buck

Converter for PMG in Low Power Applications
R. Bharanikumar1, A. Nirmal Kumar2 and K.T. Maheswari3.
1, 2, 3,
Department of Electrical and Electronics Engineering, Bannari Amman Institute of Technology, Anna University, Tamil Nadu, India
.
E-mail: [email protected], [email protected]

Abstract: This paper focuses on the design and development of 3. Small Wind Turbine in Battery Charging
synchronous buck converter in low power generation system.
The buck topology suffers from low efficiency at light loads due
Application
to dissipation that does not scale with load current. In this paper
The performance limitations of permanent magnet wind
we present a method for improving efficiency in buck converter
turbine generators in battery-charging applications are
by reducing gate drive losses. Results of PSIM simulation are
presented. caused by the poor match of the rotor, generator and load
characteristics over most of the operating wind speed range.
Keywords: Buck Converter, Synchronous Rectification, PMG, Even the small amount of energy (1kWh) that these batteries
Wind turbine. store can sufficiently improve the quality of life for such
areas, giving people access to electrical lighting, TV/radio,
1. Introduction and other household conveniences. We placed an optimizing
Small wind turbines offer a promising alternative for direct current DC/DC voltage converter between the rectifier
many remote electrical uses where there is a good wind and batteries. We can control the current output of the
resource. The goal of this work is to characterize small wind synchronous buck converter, which allows us to control the
turbines, wind-diesel hybrid system components and wind- power going to the batteries. Battery-charging systems are
hybrid systems and to develop new off-grid applications for very important in developing countries where rural families
small wind turbines in order to expand the international cannot afford a solar-battery home system or other
market for these systems. Projects fall into two electricity options. The technical aspects of charging
classifications: applications development and testing. numerous 28-V batteries with a small permanent magnet
Testing includes both small turbines and wind-hybrid alternator wind turbine suggest that a special battery-
systems. Although the projects that fall under applications charging station needs to be developed. The major
development and testing are varied, they all focus on the advantage of a centralized battery-charging station is that it
remote power market and all include small wind turbines as can bring electric service to a very low-income segment of
the power source. the population. This performance improvement comes at
higher system capital cost; however, the cost per charged
2. Block Diagram battery of the system with the individual charge controllers
is lower because of better performance characteristics.
A block diagram consists of a rectifier stage , a buck
converter and controller. Many small wind turbine
generators consist of a variable-speed rotor driving a 4. Permanent Magnet Generator
permanent-magnet synchronous generator. The principal
Permanent magnet alternators are the most powerful
application of such wind turbines is battery charging, in
and cost-effective solution for building a wind generator.
which the generator is connected through a rectifier to a
Their low-rpm performance is excellent, and at high
battery bank. The wind turbine electrical interface is
speeds they can really crank out the current due to their
essentially the same whether the turbine is part of a remote
efficiency. They provide an optimal solution for varying-
power supply for tele communications, a
speed wind turbines, of gearless or single stage gear
stand-alone residential power system or a hybrid village
configuration. [5] The evolution of the control design of
power system.
PM drives begins with the cost reduction of permanent
magnet material and follows the progress of control
theory of AC electric machinery. The main difference
between PM drives and their earlier developed counter
parts lies in the removal of the excitation field circuitry
with troublesome brushes and its replacement with
permanent magnets. But the application PM disables
classical field weakening control, because the magnets
produce constant magnetic field intensity. With the cost
reduction of rare permanent magnet materials PM
machines became very popular in industry due to their

• Simple structure
Figure 1. Block Diagram
 (IJCNS) International Journal of Computer and Network Security, 81
Vol. 2, No. 3, March 2010

• High efficiency explained below. The first state corresponds to the case
• Robustness when the switch is ON. In this state, the current through the
• High torque to size ratio. inductor rises, as the source voltage would be greater than
the output voltage, whereas the capacitor current may be in
4.1 Types of Permanent Magnet Generators either direction, depending on the inductor current and the
load current. When the inductor current rises, the energy
Modern permanent magnet generators need no separate stored in it increases. During this state, the inductor
excitation system. They can be gearless or with gearbox and acquires energy. When the switch is closed the diode is in
are fully controlled with variable speed and reactive power the off state. In Figure 2 the capacitor is getting charge.
supply. They provide the highest power quality and The second state relates to the condition when the switch
efficiency for the end user. They offer three different is OFF and the diode is ON. In this state, the inductor
concepts of permanent magnet generator technology. current freewheels through the diode and the inductor
supplies energy to the RC network at the output. The energy
4.1.1. Low Speed Robust Gearless System stored in the inductor falls in this state. In this state, the
In a direct drive application the turbine and the generator inductor discharges its energy and the capacitor current may
are integrated to form compact and structurally integrated be in either direction, depending on the inductor current and
unit. The design gives free access to all parts for easy the load current.
installation and maintenance. The simple and robust low
speed rotor design with no separate excitation or cooling
system results in minimum wear, reduced maintenance
requirements, lower life cycle costs, and a long lifetime.

4.1.2. Medium Speed Compact and Economical Unit

This is a very compact slow speed system with the
turbine main bearing and the permanent magnet generator
Figure 3. Second Stage of Basic Circuit
integrated to a single-stage gearbox giving high efficiency
with low maintenance needs. It emphasizes the same simple When the switch is open, the inductor discharges its
and robust low speed rotor design with no separate energy. When it has discharged all its energy, its current
excitation or cooling system, resulting in less wear, reduced falls to zero and tends to reverse, but the diode blocks
maintenance requirements, lower life cycle costs, and a long conduction in the reverse direction. In the third state, both
lifetime. the diode and the switch are OFF. During this state, the
capacitor discharges its energy and the inductor is at rest,
4.1.3. High Speed Small Power Pack with no energy stored in it. The inductor does not acquire
energy or discharge energy in this state.
The system is mechanically similar to the doubly fed type
with even smaller space requirements. Extremely high
power with a small size and typical speed range is from
1000 to 2000 rpm for a 6 or 8 pole generator.

The following are the key specifications evolved for PMG

PMG AC 3φ, 4-pole machine
Rated Speed 500 RPM - 3000 RPM Figure 4. Third Stage of Basic Circuit
Output Power 1.0 - 1.5 kW
Output voltage 65V Here it is assumed that the source voltage remains
constant with no ripple, and the frequency of operation is
kept fixed with a fixed duty cycle. When both the input
5. Buck Converter voltage and the output voltage are constant, the current
Output current 20A
through the inductor rises linearly when the switch is ON
and it falls linearly when the switch is OFF. Under this
condition, the current through the capacitor also varies
linearly when it is getting charged or discharged.

6. Synchronous Buck Converter

The synchronous-rectified buck converter uses current-
mode control to regulate the output voltage. This control
mode allows the converter to respond to changes in line
Figure 2. Basic Circuit of Buck Converter voltage without delay. Also, you can reduce the output
inductance to increase the converter's response to dynamic-
load conditions.
The operation of the buck converter is explained first.
This circuit can operate in any of the three states as
82 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

Although these features would appear to favor current- MOSFETs. Under light loads, the control block usually
mode control in applications that require a fast dynamic turns the lower MOSFET off to emulate a diode.
response, this control method has some disadvantages. For Synchronous rectification with discrete MOSFETs causes
example, it tends to be sensitive to noise in the control loop. variable switching delays because of the variations in gate
Also, current-mode control method requires two feedback charge and threshold voltage from one MOSFET to another.
loops: a current inner loop and a voltage outer loop, thus Standard control circuits compensate for these variations by
complicating the design. Finally, the controller uses a delaying the turn-on drive of the lower MOSFET until after
current-sensing resistor in series with the output inductor. the gate voltage of the upper MOSFET falls below a
This current-sensing resistance typically dissipates as much threshold. This delay creates a dead time in which neither
power as do the MOSFETs, further reducing the current- MOSFET conducts. The dead time eliminates the possibility
mode converter's efficiency. Voltage-mode control is of a destructive shoot-through condition in which both
attractive for low-voltage buck converters because, it MOSFETs conduct simultaneously. Standard designs use
involves a single control loop, exhibits good noise immunity the same method to delay the turn-on of the upper device. A
and allows a wide range for the PWM duty-cycle ratio. Also, typical design delays discrete MOSFET conduction with a
voltage-mode converters do not require a resistor for sensing 60-nsec dead time and limits converter switching frequency
current. However, the transfer function of standard voltage- to 300 kHz.
mode buck converters that use Schottky diodes changes from
no load to full load, making it difficult to achieve fast 6.1 Conventional Vs Synchronous Buck Converter
response to large dynamic loads. The voltage drop of a
MOSFET is much less than that of a Schottky diode, The comparison of efficiency between a synchronous
which rectifier with a parallel Schottky diode and that of a
improves the efficiency of buck converters using Schottky diode alone is shown in Figure 6.
synchronous rectification. Synchronous rectification
increases the efficiency of a buck converter by replacing the
Schottky diode with a low-side NMOSFET. The resultant
voltage drop across the MOSFET is smaller than the
forward voltage drop of the Schottky diode. A more
comprehensive comparison includes the switching losses for
both the MOSFET and the Schottky diode. However, at
typical operating frequencies and voltages, a buck
regulator's switching losses are usually small in comparison
with the conduction losses. The low-side MOSFET conducts
current in its third quadrant during the off times of the high-
side MOSFET. This synchronous switch operates in the
Figure 6. Efficiency Graph
third quadrant, because the current flows from the source to
the drain, which results in a negative bias across the switch.
A positive voltage at the gate of the device still enhances the 7. Test Results
channel. 7.1Generator Testing Details

7.1.1. G ENERATOR SPEED VS GENERATOR OUTPUT

VOLTAGE

The table shows various readings of generator speed and

their corresponding output voltages. It can be seen from the
table the generator output voltage of 51.1 volts is obtained
for the maximum speed of 1900 rpm.

Table 1: Speed and generator voltage

Figure 5. Synchronous Buck Converter Generator speed Generator o/p
As Figure 5 shows, conventional synchronous-rectified in rpm voltage
buck converters partition the PWM-control and in Volts
synchronous-drive functions into a single IC that drives 900 19.2
discrete MOSFETs. The control and driver circuits 1200 26.33
synchronize the timing of both MOSFETs with the 1350 29.41
switching frequency. The upper MOSFET conducts to 1700 46.2
transfer energy from the input, and the lower MOSFET 1900 51.1
conducts to circulate inductor current. The synchronous
PWM control block regulates the output voltage by
The following graph has been drawn for the generator
modulating the conduction intervals of the upper and lower
speed versus generator output voltage. Generator speed is
 (IJCNS) International Journal of Computer and Network Security, 83
Vol. 2, No. 3, March 2010

taken in x axis and generator output voltage is taken in y- 1200 4.10

axis. 1350 4.62
1700 5.81
1900 6.50

The following graph has been drawn for the generator

speed versus wind velocity. Generator speed is taken in x
axis and wind velocity is taken in y axis.

Generator Speed in
2000
Figure 7. Generator Speed Vs Generator Output Voltage
1500

rpm
7.1.2. Generator Output Voltage Vs Wind Velocity 1000

The table shows various readings of wind velocity and 500

their corresponding output voltages. It can be seen from the

0
table the generator output voltage of 51.1 volts is obtained 0 2 4 6 8
for the maximum wind velocity of 6.5 m/s.
Wind Velocity in m /s

Figure 9. Generator Speed Vs Wind Velocity

Table 2: Voltage with wind velocity
Generator output Wind velocity
voltage in volts in m/s
8. Simulation Results
19.2 3.07
26.33 4.10 Electronic circuit design requires accurate methods for evaluating
29.41 4.62 circuit performance. Because of enormous complexity of
46.2 5.81 modern integrated circuits, computer aided circuit analysis
is essential and can provide information about circuit
51.1 6.50
performance that is almost impossible to obtain with
laboratory prototype measurements. PSIM is a general-
The following graph has been drawn for the wind velocity purpose circuit program that simulates electronic circuits.
versus generator output voltage. Wind velocity is taken in x- PSIM can perform various analyses of electronic circuits:
axis and in y axis generator output voltage is taken. The the operating points of transistors, a time domain response,
generated voltage is maximum when the wind velocity is a small signal frequency response, and so, on. Simulation
between 6 and 12 m/s. work was done for all the circuits and results are attached.

8.1Simulated Circuit Diagram

Figure 8. Generator Output voltage Vs Wind Velocity

7.1.3. Generator Speed Vs Wind Velocity

The table shows various readings of generator speed and

their corresponding wind velocity. It can be seen from the
table the wind velocity of 6.5 m/s is obtained for the
maximum speed of 1900 rpm.

Table 3: Speed with wind velocity

Generator speed in Wind velocity in
rpm m/s
900 3.07
84 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 3, March 2010

Figure 10. Simulated Circuit Diagram of IEEE Power Electronics Conference. pp. 787-791,
2006.
[6] A.B. Raju, K.Chatterjee and B.G. Fernandes, “A
The figure shows the simulated circuit diagram. The Simple Power Point Tracker for Grid connected
Synchronous buck converter operate in current program Variable Speed Wind Energy Conversion System with
mode control .The unit is a PI controlled device that controls reduced Switch Count Power Converters”, IEEE
the power level at which the converter operates. The unit is conference on Power Electronic specillists,2003. pp
primarily designed to operate from the three-phase 456-462.
alternating current output of the wind turbine. The
following graph shows the PSIM simulation results for the
output voltage from the synchronous buck converter. Authors Profile

Bharanikumar.R was born in Tamilnadu,

India, on May 30, 1977. He received the
B.E degree in Electrical and Electronics
Engineering from Bharathiar University, in
1998. He received his M.E Power
Electronics and Drives from College of
Engineering Guindy Anna University in
2002. He has 9 yrs of teaching experience.
Currently he is working as Asst. Professor
in EEE department, Bannari Amman
Institute of Technology, Sathyamangalam, TamilNadu, India
Figure 11. Output result of Synchronous buck converter Currently he is doing research in the field of power converter for
special machines; vector controlled based synchronous machine
drives, converters for wind energy conversion systems.
9. Conclusion
Synchronous rectification is possible with all commonly A.Nirmal Kumar was born in the year
used converter topologies. It is achieved by simply adding a 1951. He completed his PG and UG in
Electrical Engineering from Kerala and
MOSFET in parallel with the free-wheeling diode. As a
Calicut University respectively. He
result of this addition, the efficiency will improve completed PhD in Power Electronics in the
significantly. A simple to operate and robust wind-electric year 1992 from P.S.G. College of
battery-charging station has been developed and tested. Technology, Coimbatore under
Future work for the applications development and testing Bharathiar University. He was with
team will include continued testing of commercial or near N.S.S. College of Engineering for nearly
commercial products for the remote electrification market. 28 years in various posts before joining Bannari
Amman Institute of Technology, Sathyamangalam, TamilNadu,
India in the year 2004. He is a recipient of Institution of Engineers
Gold Medal in the year 1989. His current research areas include
References Power converters for Wind Energy Conversion System and
Controller for Induction motor drives.
[1] Monica Chinchilla, Santiago Arnaltes, Juan Carlos
Burgos, “Control of Permanent-Magnet Generators
Applied to Variable-Speed Wind-Energy Systems Maheswari .K.T was born in Tamilnadu
connected to the Grid”, IEEE Transactions on Energy India, on Dec 20, 1980. She received her B.E
Conversion, vol. 21, no. 1, pp.130-135, , March 2006. Degree in Electrical and Electronics
[2] F. Z. Peng, “Z-Source inverter,” IEEE Trans. Ind Engineering from Erode Sengunthar
Engineering College, Erode, Bharathiyar
Applicat.,vol. 39, pp.504–510, Mar./Apr. 2003.
University. Currently she is pursuing M.E in
[3] F. Z. Peng, M. Shen, and Z. Qian, “Maximum boost Power Electronics and Drives at Bannari
control of the z- source inverter,” IEEE Transaction on Amman Institute of Technology, affiliated to
Power Electronics. vol.20, no.4, pp833-838 July2005. Anna University. Her field of interest includes PMG and
[4] Shigeo Morimoto, Hideaki Nakayama, Masayuki Sanada, Power converters.
Yoji Takeda, “Sensorless Output Maximization Control
For Variable-Speed Wind Generation System using
IPMSG” , IEEE Transactions on Industrial Applications
2003, pp.1464-1471
[5] Tomonobu Senjyu, Sathoshi Tamaki , Naomitusu
Urasaki, Katsumi Uezato Toshihisa Funabashi, Hideki
Fujita “Wind Velocity and PositionSensorless
Operation for PMSG Wing Generator”, Proceedings