International Journal of Wireless & Mobile Networks (IJWMN) Vol. 8, No.

3, June 2016

LINK-LEVEL PERFORMANCE EVALUATION OF

RELAY-BASED WIMAX NETWORK
James D. Gadze1 Griffith S. Klogo2 and Theophilus Anafo3
1,3

Department of Electrical Engineering, Kwame Nkrumah University of Science and

Technology (KNUST), Kumasi, Ghana
2
Department of Computer Engineering, Kwame Nkrumah University of Science and
Technology (KNUST), Kumasi, Ghana

ABSTRACT
Relay technology promises appreciable network throughput and coverage enhancement which is required
for high speed wireless cellular systems to function to their optimum. This paper is focused on analysing
the link performance of a relay-based WiMAX network under varying conditions. Comparison is made
between relay and direct link communication in terms of BER (Bit Error Rate), spectral efficiency and
capacity. Effect of multipath fading and user speed on performance are investigated as well. The entire
study involves a hypothetical view as well as MATLAB simulations to predict the benefit of relay
deployment. This is aimed at solving pertinent issues such as coverage holes and cell edge problems which
are associated with traditional non-relay based cellular networks.

KEYWORDS
Bit error rate, capacity, coverage enhancement, coverage hole, network throughput, relay-based WiMAX,
spectral efficiency.

1. INTRODUCTION
The deployment of wireless network technologies on the basis of the level of demand for higher
data transmission rates cannot be over emphasized. Technologies that promise to provide higher
data rates are rapidly enticing more vendors and operators. One of the most capable prospects of
such arising technologies is WiMAX (Worldwide Interoperability for Microwave Access)
described by the IEEE 802.16 standard [1]. WiMAX is intended for wireless metropolitan area
networks (WMAN) providing broadband wireless access up to 50 Km for fixed stations and 5-15
Km for mobile stations [1][2][3].
Many researchers do believe that WiMAX can propel the wireless data transmission concept to a
greater magnitude than expected. The development of the IEEE 802.16 (WiMAX) standard is as a
result of the increasing demand for higher bandwidth needed for video, voice and multimedia
communication. This has impacted immensely on the manner in which people interact or
communicate as well as enjoy their entertainment.
A traditional WiMAX cellular network earmarks a BS (base station) to offer services within a
given radius of coverage. The core limitation of this architecture include low spectral efficiency at
cell boundaries as a consequence of low SNR (signal to noise ratio) and coverage holes as a result
of shadowing caused by land forms in mountainous areas and terrestrial obstacles such as high
rise buildings in urbanized settings as depicted in figure 1.1. These factors impede drastically on
signal transmission and deplete coverage to users in shadowed and far-fetched regions resulting in
low data rates.
DOI: 10.5121/ijwmn.2016.8303

35

International Journal of Wireless & Mobile Networks (IJWMN) Vol. 8, No. 3, June 2016

Figure 1.1: Conventional cellular coverage problems

Cellular problems of these nature have received much interests from both academic and industry
researchers with some countermeasures proposed in aspects of interference management such as
OFDM, MC-CDMA, etc., and also cooperation techniques such as MIMO and smart antennas.
However, Spectrum efficiency near the cell edge is still poor as users suffer from enormous path
loss and large inter-cell interference. Users require increased transmit power due to low bit energy
to noise ratio at these regions in order to maintain bits fidelity and good throughput [3]. These
prevailing issues contribute to making it almost impossible to achieve the set target of services
anywhere, anytime as subscribed in the objectives of IEEE 802.16 WiMAX standard.
A simple solution to solving these problems is to strategically increase the density of BS within a
geographical area which could also extend coverage to far-reached places. This approach is
however cost ineffective to network operators and inefficient when few users are to be served.
The use of a simple form of a BS called relay station (RS) is the rapid and cost effective way of
deploying the network infrastructure [4]. Relays do not require E1 or T1 backhaul connection to
communicate with BS that has link connection with some portion of their air link bandwidth [3].
User information from a nearby mobile station (MS) or user can be forwarded to a base station
with the help of the relay station as shown in figure 1.2. Signal coverage can be effectively
extended while enhancing overall throughput via the use of relay.

Figure 1.2: Typical Relay WiMAX Network

Types of relays that can be deployed include NT-RS (non-transparent relay station) and T-RS
(transparent relay station) [5]. NT-RS are usually deployed to offer coverage at cell edge or
beyond the coverage of a BS. This offer total connectivity interface between the BS and MS
where traffic signal together with control signals and preamble are transmitted through relay. The
other type; T-RS is deployed within BS coverage area to relay only traffic signal between BS and
MS. BS send control information to MS directly in this mode. T-RS are basically used to
improve throughput and capacity within the cell [6].
36

International Journal of Wireless & Mobile Networks (IJWMN) Vol. 8, No. 3, June 2016

In this paper, we investigate the link-level performance of communication through relay and
compare with direct link communication in a single WiMAX cell scenario in a mobile
environment. Link-level performance evaluation is significant to the assessment of the behavior
of a single communication link under varying condition which could provide essential results
necessary to estimate the potential benefit of utilizing relays in solving the issues of coverage
holes and reduced spectral efficiency at cell boundaries. A study in this direction will inform the
choice and proper deployment of relays which will enhance coverage and overall cell throughput.
The rest of the paper is organized as follows: Section II presents the characteristics of a relaybased network. Section III presents the some relay protocols. In section IV, link-level simulation
is performed to investigate link performance calculation of BER, spectral efficiency and capacity.
Section V summarizes the results of the thesis and informs a potential direction leading to future
research.

2. CHARACTERISTICS OF RELAY-BASED NETWORKS

Relay-based wireless network is a multi-hop design having a fixed infrastructure in which a Relay
Station (RS) forms a different entity besides the BS and its associated Mobile Stations (MSs).
Typically, a few low cost relays and their associated BSs form a relay-based system. Relay
systems are mostly considered applicable in the network operator contexts as the operator plans
and deploys the wireless access network. These settings are naturally characterised by a treebased routing setup in which end user terminals connect to BS via short two to three hops routes.
The IEEE 802.16j-2009 Relay WiMAX is a typical example of this architecture. Authors in [7][8]
introduced the basic theories and principles of relay WiMAX presenting most of the major
modifications in the IEEE802.16j standard. While authors in [9] and [10] have detailed the key
concept and benefits of this new network design providing motivation and features relating to
relay-based system design. They particularly indicate throughput enhancement, coverage
enhancement and deployment cost as the three main benefits derived from relay-based
architecture compared to conventional single-hop system. Each of these key benefits is further
detailed below.

2.1. Throughput Enhancement

Relay design is expected to increase the throughput for users at the cell edge of a WiMAX
system. Notably, subscribers at cell edges may be required to communicate at reduced rates [11]
due to low received signal strength at these regions. In addition, relay-based design is expected to
increase system capacity by deploying relays such that it enables more aggressive spatial re-use.

2.2. Coverage Extension

The use of a relay is probable to improve coverage in severely shadowed regions as well as
extend the coverage range of a BS. In either case, coverage is improved via the RS transmitting
from an advantageous location nearer to a disadvantaged user [12][13].

2.3. Cost Reduction

There is a potential cost gains in relay based networks deployment than single hop wireless
systems. Relay stations are relatively cost effective than base stations. However, wider coverage
and enhanced throughput can be achieved at low cost by adding relays in a network which is
reasonable than deploying only expensive base stations in order to offer good coverage and
enhanced capacity [5]. Relays which have lower complexity compared to BSs are expected with
no wired connection to the backhaul [9] which is significant in providing a good cost performance
in deploying a cellular network.
37

International Journal of Wireless & Mobile Networks (IJWMN) Vol. 8, No. 3, June 2016

3. RELAYS PROTOCOLS
Relay protocols are classified based on the signal processing techniques they employ. Authors in
[14] proposed two main techniques namely, Amplify and Forward (AAF) and Decode and
Forward (DAF) which are eminent to process the received signal at the relays. Each technique is
described below.

3.1. Amplify-and-Forward
AAF protocol is mostly used in analog repeaters which simply amplifies and forwards the
received signal in the downlink and uplink. This method offer an advantage in terms of delay
however, certain issues arise including: increase in system noise level, increase in interference
level due to lack of specific scheduling and erroneous data is forwarded without correction over
multiple links [14].

3.2. Decode-and-Forward Scheme

In DAF scheme, received signal at the relays are fully decoded and re-encoded before signal is
retransmitted which offer some advantages for the management of radio resource. Resource
allocations to both relay and access links becomes quite easier making this scheme a worthy
candidate for OFDMA-based systems. Additionally, this scheme is appropriate in environment
where heterogeneous modulation and coding schemes are used such as in OFDMA systems.
However, this scheme define some latency issues as some amount of processing time is required
for the signal decoding and re-encoding process [14].

3.3. Related Work

There has been a substantial work available in literature on relay protocols many of which are on
analytical performance comparison of Amplify and Forward (AAF), Decode and Forward (DAF)
and other relay protocols.
In [15], authors considered both AAF and DAF relay schemes investigating outdoor-to-indoor
relay communication links and quantifying the end-to-end channel capacity. In measuring the
links constituting the relay transmission, Real-time sounder was utilized. They established the
extent to which capacity is improved when DAF relay scheme is used compared to when AAF is
utilized. However, they did not establish the effect of latency and resource consumption in their
model.
In [16], DAF Relay is used to allow for opportunistic relay of packets which helped to propagate
redundant signals over multiple paths. The authors utilised this approach to combat the channel
noise and fading effects. They investigated the use of packet combining in cooperative diversity
scenario which exploits the inherent spatial diversity along with the inherent time diversity in
Automatic Receive Request (ARQ) retransmissions. Additionally, their approach sought to save
resource by relaying packets only at instants when direct transmission fails with performance
measures in BER and throughput in AWGN and also in multipath fading.
System level performance simulation is employed in [17] with a simple AAF relay protocol
operating in full duplex mode in a multihop environment. The authors investigated bandwidth
efficiency and the associated service outage performance for different relay scenarios which
aimed at improving carrier-to-interference ratio in heavily shadowed region. The issue with their
system is the minimal throughput gain experienced by relays in open space due to co-channel
38

International Journal of Wireless & Mobile Networks (IJWMN) Vol. 8, No. 3, June 2016

interference from relays in neighbouring cells. Meanwhile, increased system capacity was
demonstrated by fully reusing the frequency in all relays while improving the per-user data rate in
cell edges and heavily shadowed regions. Authors in [18] investigated the performances for fullduplex in two-way AAF relay channel consisting of two source nodes and a single relay node in
terms of outage probability and ergodic capacity. They were able to establish better performance
in the presence of loop interference in full duplex mode than in two non-orthogonal half duplex
channels.
A situation of power allocation at relays was addressed in [19] to increase the transmission rate in
a non-orthogonal AAF relay network where amplified signals are simultaneously transmitted at
the same time and with the same frequency. A unified power allocation in AAF approach is
presented in [20] in multi-hop OFDM relaying systems. Authors considered a short and long-term
individual and total power constraint at the BS and RSs, and formulate decentralized low
complexity power allocation algorithms when links are subjected to fading. Two stages were
adopted, including a power distribution phase among distinct subcarriers, and a power allocation
phase among different relays.

4. CHALLENGES IN PLANNING WIMAX CELLULAR RELAY-BASED

NETWORKS
Network planning is quite a challenging task in rolling out a cellular network. These challenges
come in many aspects and the incorporation of relays into the network gives rise to new issues.
To ensure that the installed network is enhanced, it is necessary to investigate and determine
certain facets in relation to the number of relays to deploy along with their location in order to
address the relay selection process while mitigating the interference through frequency re-use.

4.1. Relay Placement and Transmission

Relay stations (RS) are the nodes found between the BSs and MSs thereby reducing the
transmission distance. System performance is sensitive with respect to the number of deployed
relays as well as their location. If well placed could enhance system capacity, system reliability as
well as throughput per user. The problem related to relay placement has been studied in different
wireless multi-hop networks such as in WLANs [21] and sensor networks [22]. The authors in
[23] suggested an approach to maximise the overall system capacity by defining an algorithm for
relay placement and also determining the number of relays to deploy. They clearly demonstrated
significant gains via careful relay deployment. However, their work also demonstrated that no
additional gain is achieved above a certain number of relays deployed. Therefore, defining the
relay position under certain realistic assumptions is presented and perceived as NP-hard problem
and is hence a challenging problem to solve [24]. However, if the MS distribution is assumed to
be uniform in an ideal scenario according to [5], system optimization can be achieved by adopting
an iterative algorithm in the placement of RS. In practice, non-uniform distributed traffic demand
must be considered especially in large geographic areas. In the first stage of network deployment,
the BSs locations are first determined followed by the design of the RS location algorithm to
locate the RSs [4].
The IEEE 802.16j mobile multi-hop relay network utilizes several transmission schemes for its
relays as surveyed by authors in [4][25] as captured in the standard [5]. Authors is
[26][27][28][29] present some other studies on relay placement which focus on finding the best
location for RS in order to improve performance by extending the coverage of the WiMAX
system. In [26], the authors established a clear relationship between relay bandwidth reservation
and the placement of the relay station. They also recommended an optimisation scheme where
relay bandwidth reservation and relay placement are combined.
39

International Journal of Wireless & Mobile Networks (IJWMN) Vol. 8, No. 3, June 2016

4.2. Frequency Re-use

Radio frequency is a limited, scarce and expensive resource. It is largely improbable for same
frequency to be used in adjacent cells. Frequency re-use is a mechanism that promises efficient
utilisation of limited radio resource [30]. This technique has attracted the interest of several
researchers in the industry. Authors in [31] gave comprehensive overview of the basic mechanism
and algorithms involved in frequency planning and re-use processes for single-hop network. In
[32], different re-use factors for cellular multi-hop network scenario are investigated and their
respective impact on the system capacity. They demonstrated that via the use of directional
antennas to limit the interference on the relay links, relay-based systems can offer significant
gains over conventional single-hop systems.

4.3. Routing Management

Routing procedure is a significant aspect of Relay WiMAX which impact immensely on
throughput. The path selection is very critical as there is a cost associated with it in terms of
throughput and delay which warrant for an intelligent routing algorithm to ensure good
performance of the system.
However, several routing algorithms are found in literature with different measures used to
analyse the performances of different paths. Authors in [33] introduce a metric which constitutes
the bandwidth unit needed to transmit a fixed amount of data by using a specific Modulation and
coding scheme. The efficiency of the routed path used to transmit a bit over a medium by their
metric does not account for the traffic load at the relay. Authors in [34] presented a routing
scheme in multi-hop Relay-based WiMAX. Their proposed path selection maximizes the
throughput of the network while minimizing latency. In finding the best route, the authors
considered the Signal to Noise Ratio, the number of hops and the available link bandwidth in their
decision. The achievable throughput is considered by authors in [33] and [35] by reckoning the
bandwidth at the relay in terms of available free slots. This quite profound in managing the traffic
load in QoS oriented wireless system. However, how to determine the available slots at each relay
is not discussed. Additionally, defining the amount of resources available at the relay is directly
related to the scheduling algorithm and the way the resources are managed; either centralised or
distributed mode. Further in their bandwidth-based metric scheme, they did not investigate the
impact of the additional signalling overhead associated with their approach. Authors in [36]
conducted a system level performance in transparent relay mode demonstrating the fact that, there
can only be up to 10% increase in maximum throughput in the downlink. They also argued in the
article that throughput enhancement can only benefit a half of the BS coverage. Routing
procedures in Relay WiMAX systems differ depending on the manner resources are managed.
Relay WiMAX routing issues seem less complex than in other wireless systems such as wireless
mesh networks. However, there exist some pertinent issues in relation to path selection
specifically with non-transparent relay systems.

4.4. Resource Allocation

Relay WiMAX significantly has an increases signalling overhead compared to the conventional
single-hop WiMAX system. Therefore, to meet the different QoS requirements of each user
require the design of novel scheduling algorithms as well as carefully managed signalling for an
efficient resource allocation.
Authors in [37] and [38] analysed the MAC efficiency issues with Relay WiMAX proposing
different schemes for the MAC efficiency optimization. A new downlink resource allocation
algorithm is proposed in [37] which dynamically allocate resources on the downlink for relay
40

International Journal of Wireless & Mobile Networks (IJWMN) Vol. 8, No. 3, June 2016

transmission with respect to bandwidth requirement. However, this work did not consider
OFDMA and scheduling not covered at all. Authors in [38] sought to reduce system overhead
while improving the efficiency of utilisation of the MAC frame by proposing an aggregation
scheme. The need for connection and packet aggregation which combat significant overhead was
highlighted in this work. But notably, the Relay WiMAX standard different schemes from those
described in [38] are defined to aggregate the relay link traffic.
Generally, most resource allocation work evolving through the years focus on the MAC
efficiency optimization and does not follow the Relay WiMAX specification. These works
significantly rely on system-level performance.
One aspect which is imperative with multiple relays deployment is the possibility to enable
concurrent transmission at the relays which is significant for efficient utilisation of available radio
resources provided interference is well managed. Resource scheduling in an urban setting under
severe shadowing was investigated by the authors in [38]. They assumed in their system 4
directional antennas on both the base station and the relay station which resulted in their
simulation highlighting on the advantages of directional antennas over omni-directional antennas
in an urban settings. Notably, system throughput can be increased by 6 or 12 times with the
proposed method compared to an omni-directional system. Additionally, this work greatly utilised
spatial re-use for throughput enhancement which is further spot on via simulation studies in
[40][41]. The authors emphasised the need for distributed scheduling (i.e. scheduling at the BS as
well as RSs) which maximises concurrent transmissions while preventing severe interference due
to wrong re-use decisions. However, the frame structure considered in these works does not
conform to Relay WiMAX specifications and the relay type considered is not clear.
In [42], authors discussed several publications focusing on the management of radio resource in
OFDM relay based systems which can be adapted by Relay WiMAX systems. Various
approaches for dynamic resource allocation found in literature were presented.
Several of these proposals on system performance enhancement exhibit some short fall either by
their signalling requirements or their complexity in implementation. Distributed scheduling is one
solution discussed by authors in [42] which promises a reduced overhead and system complexity.
Largely, this review demonstrates that there is still much work needed in the performance
enhancement process of OFDM relay systems and hence relay WiMAX systems. Relay WiMAX
systems requires a scheduling algorithm design which takes into account the detailed issues with
regards to the frame structure in transparent or non-transparent mode, signalling and QoS.
Although some clues are offered in these works with respect to achievable throughput gain of
relay WiMAX systems, a lot more work is still left aloof and worth considering in terms of the
best design practises and performance measures in different scenarios.

4.5. Summary
The Relay WiMAX is relatively a new technology with a new architecture which presents several
complexities amidst an already thought-provoking wireless environment in a mobile world.
Largely, system design becomes difficult due the extensive aggregate of choice available to
engineers tied with a general deficiency of understanding of the different design decisions and
their impact.

41

International Journal of Wireless & Mobile Networks (IJWMN) Vol. 8, No. 3, June 2016

The performance enhancement offered by Relay WiMAX systems over single hop WiMAX
systems is influenced by numerous factors including network topology, relay mode, transmit
power level at the relays and the type of antennas used. A profound understanding of the
relationship between the enhancement of relay based WiMAX systems and the multiple
parameters that are necessary to the system design can offer a roadmap for the deployment of
relay-based WiMAX systems. The focus of the reviewed work provided is to give a better
understanding of the impact of different design options on performance.
Several issues pertaining to Relay WiMAX remain unanswered as several research works are
based on inaccurate assumptions. Hence, the need for much work to be done considering different
system configurations which is essential in determining the most practicable scenarios of this
technology. It is imperative to note that, transparent and non-transparent systems produce
different issues and different approaches are required to solve them. While the reviewed literature
in this section are essential and complement this paper naturally, some significant system
configuration components that impact on performance are not considered.

5. SYSTEM MODEL
Several assumptions were made due to system complexity. We considered two hops with two
kinds of users, a direct link user or MS (mobile station), communicating directly with the BS and
relay link MS, communicating through relay stations to the BS as depicted in fig. 3. This is due to
the fact that more than two-hop relaying without extending coverage reduces the efficiency of
using RSs.

Figure 5.1: System model

The focus is on only the downlink communication whiles considering transparent relay with AAF
(Amplify and Forward) relay protocol. Every node has a single omni-directional antenna and
operates in half-duplex mode; hence, no terminal can receive and transmit data simultaneously.
For multipath fading, Rician fading and Rayleigh fading were considered for BS-RS link and
access link respectively. For channel propagation models, SUI channel model and Cost 231 Hata
model were used for BS-RS link and access link respectively.
The BS is responsible for allocating resources and is connected to the backhaul network, while
the RS has no backhaul link but is wirelessly connected to the BS. The main responsibility of the
RS is to relay data between the BS and MS. The one-hop links; BS to RS and RS to RS are
referred to as relay links, and users served through these links are classified as relay link users.
Similarly, the links; BS to MS are referred to as direct links and users served directly by the BS
are considered direct link users.
42

International Journal of Wireless & Mobile Networks (IJWMN) Vol. 8, No. 3, June 2016

5.1. Channel
In order to model such wireless channels, we can suppose that for each point in the 3-dimensional
space, the wireless channel is a linear time-varying filter with following impulse response [49]
N 1

h (t , ) = ck (t ) ( k (t ))e j k ( t )

(1)

k =0

where t and are the observation time and application time of the impulse, respectively.
is the number of multipath components, k( ), k( ), k( ) are the random time - varying
amplitude, arrival-time, and phase sequence, respectively, and is the delta function. The output
of the wireless channel for a given transmitted signal s( ) can be calculated by:

s(t )h(t, )d + n(t )

y (t ) =

(2)

where ( ) is the background noise.

5.2. Multipath Fading Channels

We assume well known fading channels such as the Rayleigh fading for the access links (i.e. BSMS and RS-MS links) and Rician fading channel for the relay link (i.e. BS-RS link). The
Rayleigh fading channel is most applicable when there is no propagation along the line of sight
between the transmitter and receiver, while the Rician fading channel is more appropriate when
there is a dominant line of sight component at the receiver.
For BS-MS and RS-MS access links, we can precisely model via a Rayleigh distribution the
amplitude distribution of the received signal as:

p( ) =

exp
2 , 0
2
2

(3)

Where represent the received signal amplitude and 2 represent the local mean power [50].
Now according to the amplitude of the received signal probability density function (pdf) in (3),
we can derive the received signal power pdf, , which has the exponential pdf:

p( ) =

exp , 0

*
1

(4)

where * represent the received signal mean power. Similarly, for the relay (BS-RS) link, we can
via Rician distribution model the amplitude distribution of the received signal as:

p( ) =

2 + v2 v

exp

I0 2 , 0, v 0
2
2
2

(5)

1 2
v
If In(.) represent a modified n -order Bessel function of the first kind, 2
characterise the
2
power of the LOS component while the power of all other scattered components denoted as ;
th

43

International Journal of Wireless & Mobile Networks (IJWMN) Vol. 8, No. 3, June 2016
*
we can deduce the total mean power of the received signal as:

1
2

(6)

* = v2 + 2

The Rician K-factor is given as the ratio between the signal power in the dominant component

K=

v2
2 2 . The Rician distribution becomes

and the local mean scattered power [51], where

Rayleigh when the K-factor goes to zero; the pdf of , (received signal power) from equation (5),
can be obtained by transforming random variable into via the amplitude and power relation of

=
the signal

p( ) =

1 2

2 . According to [69] the pdf of received signal power is given by:

(1 + K )e K

(1 + K )
exp
*

4 K (1 + K )
I 0
*

, K 0, 0

(7)

*
Where is the received signal mean power.

5.3. Relay Strategy

An MS located within BS coverage can either be served directly by BS or via RS. More often
than not, the MSs will prefer an RS route only if the achievable relay link data rate (BS-RS-SS) is
superior to the direct link data rate (BS-SS). The relay link data rate is influenced by the link
capacities and time durations of the linking two hops. The assumption made in our case is that
every node has a single antenna and operates in half duplex mode and also, data arriving equals
data leaving at the RS node.
For each transmission time interval according to [5], the BS allocates sub-channels to both relay
and direct links. Within a segment, if the kth sub-channel represented by ak,m is allocated to a
direct link user m then;

1 , subchannel k allocated
ak ,m =
otherwise
o ,

(8)

Likewise

bk ,r , m

1 , subchannel k allocated to user using RSr

=
otherwise
0 ,

(9)

Where bk,r,m represent kth subchannel allocated to user m using RSr (relay station r) using the
second hop link [52].
Estimating the capacity of direct link in the first time slot using Shannons formula is given by:

Ck ,m = Bk log 2 (1 + pk ,m gk ,m 2 )

(10)

With Bk representing the bandwidth of the each channel, Pk,m represent the transmit power on the
kth subchannel of BS to MS m, gk,m represent the gain of subchannel and 2 represent the AWGN
power level [53].
44

International Journal of Wireless & Mobile Networks (IJWMN) Vol. 8, No. 3, June 2016

Similarly capacity of the relay link is

Ck ,r = Bk log 2 (1 + pk ,r gk ,r 2 )
Where

Pk , r

(11)

denotes the transmit power on the

kth

subchannel of BS to RS r.

for m number of users on the relay link, the above equation becomes:

Ck ,r,m = Bk log 2 (1 + pk ,r,m g k ,r ,m 2 )

(12)

On the direct link, the realistic data rate in the first time slot according to [53] is given by:
K

C m = Ck , m a k , m , m D

(13)

k =1

Where

D represent a set of direct link users.

And on the relay link, in the second time subslot, data rate for user m is given by:
K

Cm( 2 ) = Ck ,r ,mbk ,r ,m ,

m R

(14)

k =1

In the first time slot, all relay link users share the data rate of RS r, therefore producing a data rate
in the first time subslot for each user in the relay link as:

Cm(1) = Cr(1) .Cm( 2)

(1)

If

Cr

Cm( 2 )

(15)

Rr

denotes data rate of RS r of the first hop link given by:

(16)

Cr( ) = Ck ,r ak ,r
1

k =1

If a single user is considered as in our scenario, then

Cm1 = Cr1 .

Now, given that tBS-RS represent the first time subslot and tRS-MS represent the second time subslot;
the total duration of the downlink subframe transmission should be tBS-RS + tRS-MS. The amount of
data transfer from BS to RS is equal to that from RS to MS.

Cm1 .tBS RS = Cm2 .t RS MS

(17)

To determine the average data rate of user on the relay link, we divide the amount of data
received by the time required to receive it [52].

45

International Journal of Wireless & Mobile Networks (IJWMN) Vol. 8, No. 3, June 2016

Cm =

Cm1 .tBS RS
tBS RS + tRS MS

(18)

Since RS cannot receive from BS while transmitting to MS. Subsequently, using the relation in
[53], we can rewrite the relay data rate of the MS as:

1
1
1
= 1 + 2
C m Cm Cm

(19)

6. SIMULATION RESULTS AND DISCUSSION

In the simulation setup, a WiMAX cell scenario of BS is considered and RS deployed within its
coverage. A downlink transmission is considered with a channel bandwidth of 5MHz and a
2.5GHz carrier frequency. The BS is centralised and the RS kept in amplify and forward (AAF)
mode.
First, we estimate the link condition in terms of BER between the BS, the RS and the MS as a
function of bit energy to noise ratio (Eb/N0) and subsequently spectral efficiency and the link
capacities are also evaluated. We used the SUI channel model for link between BS-RS and Cost
231 channel model between BS-MS and RS-MS links while specifying the velocity of MS within
the Doppler frequency. Initialization parameters and input data form the core of the program. At
the time of initialization, parameters that can be set include the number of subcarriers, the
nominal bandwidth, CP length, symbol modulation level and coding rate, range of the bit energy
to noise ratio (Eb/No) values (from 0 to 40 dB) and channel model parameters for simulation. The
input data stream is randomly generated. Output variables are available in MATLAB workspace
while bit error rate (BER) values for different Eb/No are stored in text files which facilitate in the
plots. In our model, the simulation of the system is repeated and the number of transmitted bits
and bit errors are calculated for each simulation. At the end, BER is estimated as the ratio of the
total number of observed errors and the total number of transmitted bits.
The performance of the system model is evaluated using varied modulation and coding schemes
such as BPSK , QPSK , QPSK , 16-QAM , 16-QAM , 64-QAM 2/3 and 64-QAM
while considering several MS velocities.
Our simulations can be summarised into 3 stages:
1. BER and Spectral efficiencies at fixed speed of MS in pure AWGN channel.
2. BER at varied speed of MS in fading channel (i.e. Rayleigh and Rician) compared to that
of non-fading pure AWGN channel.
3. Capacities of the two links under study (i.e. relay and direct link).
The next sections present a set of plots to identify trends in signal quality as we change different
parameters. These plots include BER vrs. Eb/N0, spectral efficiency vrs Eb/N0 and capacity vrs
Eb/N0 plots.

6.1. Performance measure of Relay and Direct links in AWGN channel

In this section, we have presented various BER vrs Eb/N0 and spectral efficiency vrs Eb/N0 plots
for all the mandatory modulation and coding schemes as specified in the IEEE 802.16j standard
for both relay and direct link. Figures 6.1 6.14 show these plots exhausting BPSK, QPSK, 16QAM and 64-QAM in pure AWGN using OFDM comparing performances of the two links. For
each cited modulation, the BER of relay and direct links are compared as well as their respective
spectral efficiencies estimated.
46

International Journal of Wireless & Mobile Networks (IJWMN) Vol. 8, No. 3, June 2016

BPSK (Code Rate= 1/2), Velocity = 60km/h

QPSK, (Code Rate= 3/4), Velocity = 60km/h

10

10

Relay Link
Direct Link

Relay Link
Direct Link
-1

10
-1

BER

BER

10

-2

10

-2

10

-3

10

-4

10

-3

10

6
8
Eb/No(dB)

10

12

6
8
Eb/No(dB)

14

Figure 6.1: BER against Eb/N0 for MS at 60km/h

via AWGN channel (BPSK-1/2)

10

12

Figure 6.5: BER against Eb/N0 for MS at 60km/h

via AWGN channel (QPSK-3/4)
Spectral Efficiency vs. Eb/No
QPSK, (Code Rate= 3/4), Velocity = 60km/h

Spectral Efficiency vs. Eb/No

BPSK (Code Rate= 1/2), Velocity = 60km/h

1.5
Relay Link
Direct Link

0.5
Relay Link
Direct Link

0.45

Spectral Efficiency (b/s/Hz)

0.4
Spectral Efficiency (b/s/Hz)

14

0.35
0.3
0.25
0.2
0.15

0.5

0.1
0.05

0
0

10

15

20

25
30
Eb/No(dB)

35

40

45

Figure 6.2: Spectral Efficiency against Eb/N0 for

MS at 60km/h via AWGN (BPSK-1/2)

10

15

20
Eb/No(dB)

25

30

35

40

Figure 6.6: Spectral Efficiency against Eb/N0 for

MS at 60km/h via AWGN (QPSK-3/4)
16-QAM, (Code Rate= 1/2), Velocity = 60km/h

QPSK, (Code Rate= 1/2), Velocity 60km/h

50

10

10

Relay Link
Direct Link

Relay Link
Direct Link

-1

-1

10

-2

10

10

-2

BER

BER

10

-3

-3

10

-4

10

10

-4

10

-5

10

-5

10

6
8
Eb/No(dB)

10

12

Figure 6.3: BER against Eb/N0 for MS at 60km/h

via AWGN channel (QPSK-1/2)

15

20

25

Spectral Efficiency vs. Eb/No

16-QAM, (Code Rate= 1/2), Velocity = 60km/h
2

Relay Link
Direct Link

Relay Link
Direct Link

1.8
1.6
Spectral Efficiency (b/s/Hz)

0.8
Spectral Efficiency (b/s/Hz)

10

Figure 6.7: BER against Eb/N0 for MS at 60km/h

via AWGN channel (16QAM-1/2)

Spectral Efficiency vs. Eb/No

QPSK, (Code Rate= 1/2), Velocity 60km/h

0.7
0.6
0.5
0.4
0.3

1.4
1.2
1
0.8
0.6
0.4

0.2

0.2

0.1
0

Eb/No(dB)

1
0.9

14

0
0

10

15

20
25
Eb/No(dB)

30

35

40

45

Figure 6.4: Spectral Efficiency against Eb/N0 for

MS at 60km/h via AWGN (QPSK-1/2)

10

15

20
Eb/No(dB)

25

30

35

40

Figure 6.8: Spectral Efficiency against Eb/N0 for

MS at 60km/h via AWGN (16-QAM-1/2)

47

International Journal of Wireless & Mobile Networks (IJWMN) Vol. 8, No. 3, June 2016

Spectral Efficiency vs. Eb/No

64-QAM ,(Code Rate= 2/3), Velocity = 60km/h

16-QAM ,(Code Rate= 3/4), Velocity = 60km/h

10

Relay Link
Direct Link

Relay Link
Direct Link

3.5

-1

Spectral Efficiency (b/s/Hz)

10

-2

BER

10

-3

10

-4

3
2.5
2
1.5
1
0.5

10

10

15

-5

10

10

15

20

25

Eb/No(dB)

Figure 6.9: BER against Eb/N0 for MS at 60km/h

via AWGN channel (16QAM-3/4)

20
Eb/No(dB)

25

30

35

40

Figure 6.12: Spectral Efficiency against Eb/N0 for

MS at 60km/h via AWGN (64-QAM-2/3)

Spectral Efficiency vs. Eb/No

16-QAM ,(Code Rate= 3/4), Velocity = 60km/h

64-QAM , (Code Rate= 3/4), Velocity =60km/h

10

Relay Link
Direct Link

Relay Link
Direct Link

-1

10

BER

Spectral Efficiency (b/s/Hz)

2.5

1.5

-2

10

0.5

10

15

20
Eb/No(dB)

25

30

35

40

Figure 6.10: Spectral Efficiency against Eb/N0

for MS at 60km/h via AWGN (16-QAM-3/4)

-3

10

10

15

20
Eb/No(dB)

25

30

35

40

Figure 6.13: BER against Eb/N0 for MS at

60km/h via AWGN channel (64-QAM-3/4)

64-QAM ,(Code Rate= 2/3), Velocity = 60km/h

10

Relay Link
Direct Link

Spectral Efficiency vs. Eb/No

64-QAM , (Code Rate= 3/4), Velocity =60km/h
4.5

-1

10

Spectral Efficiency (b/s/Hz)

BER

3.5
-2

10

-3

10

3
2.5
2
1.5
1

-4

10

10
12
Eb/No(dB)

14

16

18

20

0.5

Figure 6.11: BER against Eb/N0 for MS at

60km/h via AWGN channel (64QAM-2/3)

10

15

20
Eb/No(dB)

25

30

35

40

Figure 6.14: Spectral Efficiency against Eb/N0 for

MS at 60km/h via AWGN (64-QAM-3/4

We observed that, the relay link marginally performed better than the direct link at an MS
velocity of 60km/h. It was also prominent that BPSK is more power efficient and require less
bandwidth than all the other higher modulation techniques used. In case of bandwidth utilization,
the 64-QAM modulation requires higher bandwidth but gives excellent data rates as compared to
the lower modulations while QPSK and 16-QAM performances fall intermediate between the two
extremes and need higher bandwidth and are less power efficient than BPSK. But they required
lesser bandwidth and lower data rates than 64-QAM. BPSK has the lowest BER while the 64QAM has the highest BER.

48

International Journal of Wireless & Mobile Networks (IJWMN) Vol. 8, No. 3, June 2016

6.2. Effect of speed and fading on Performance

The results presented here are for simulations performed for the MS velocities of 30, 60 and 90
km/h in fading and non-fading (pure AWGN) channels. The prime purpose of this section is in
two folds. The first is to investigate the effect of speed to system performance and the next is to
probe into the influence of fading channel to performance. Figures 6.15, 6.16 and 6.17 present the
BER curves of QPSK-1/2, 16-QAM-1/2 and 64-QAM-2/3 with varied MS speed of 30, 60 and
90km/h while considering a relay link in pure AWGN channel. Figures 6.18, 6.19 and 6.20
present BER curves with same parameters in multipath fading channel.
It is clearly indicated from the results obtained that fading channels offer degraded performance
compared to non-fading channels. On the side of testing, speed has a great impact on
performance. It was also well noted that better performance is experienced when MS is moving at
slower speed than when moving faster.
QPSK, (Code Rate= 1/2)

16-QAM, (Code Rate= 1/2),

10

10
30km/h
60km/h
90km/h

30km/h
60km/h
90km/h

-1

-1

10

-2

BER

BER

10

10

-3

-3

10

10

-4

10

-2

10

-4

10

15

20

10

25

10

15

20
Eb/No(dB)

Eb/No(dB)

Figure 6.15: BER against Eb/N0 for QPSK-1/2 in

AWGN, V=30,60 and 90 km/h

30

35

40

Figure 6.17: BER against Eb/N0 for 16-QAM-1/2

in AWGN, V=30,60 and 90 km/h

QPSK, (Code Rate= 1/2)

25

16-QAM, (Code Rate= 1/2),

10

10

30km/h
60km/h
90km/h

60km/h
30km/h
90km/h

-1

10

-1

-2

10

BER

BER

10

-2

10

-3

10

-4

10

10

15

20

25

Eb/No(dB)

Figure 6.16: BER against Eb/N0 for QPSK-1/2 in

Fading channel, V=30,60 and 90 km/h

-3

10

10

15

20

25

Eb/No(dB)

Figure 6.18: BER against Eb/N0 for 16-QAM1/2 in Fading channel, V=30,60 and 90 km/h

49

International Journal of Wireless & Mobile Networks (IJWMN) Vol. 8, No. 3, June 2016

64-QAM ,(Code Rate= 2/3)

64-QAM ,(Code Rate= 2/3)

10

10
30km/h
60km/h
90km/h

-1

10

30km/h
60km/h
90km/h
-1

10
-2

BER

BER

10

-3

-2

10

10

-3

-4

10

10

-5

10

10

15

20

-4

25

10

Eb/No(dB)

Figure 6.19: BER against Eb/N0 for 64-QAM-2/3

in AWGN, V=30,60 and 90 km/h

10

15

20
Eb/No(dB)

25

30

35

40

Figure 6.20: BER against Eb/N0 for 64-QAM-1/2

in Fading channel, V=30,60 and 90 km/h

6.3. Capacities of relay and direct links

This sub-section focused on link capacities in AWGN and multipath fading channels. Figure 6.21,
presents link capacities of relay and direct links in pure AWGN non-fading channel while figure
6.22, shows the link capacities in multipath fading channel. From our results obtained, it is
observed that relay links offer higher capacity compared to direct links. However, both link
capacities reduce marginally in multipath fading channels.
Capacity of Links in AWGN Channel

Capacity of Links in Multipath Fading Channel

5

7
Relay Link
Direct Link

Relay Link
Direct Link

4.5
4
3.5
Capacity (bps/Hz)

Capacity (bps/Hz)

3
2.5
2
1.5

2
1
0.5

0
-10

0
-10

-5

5
Eb/No (dB)

10

15

Figure 6.21: Link Capacities of Relay and Direct

links against Eb/N0 in pure AWGN channel

20

-5

5
Eb/No (dB)

10

15

20

Figure 6.22: Link Capacities of Relay and Direct

links against Eb/N0 in pure Multipath Fading
channel

7. CONCLUSIONS
The results show how various network parameters such as relay deployment, modulation and
coding profiles, multipath fading and speed of users affect link performance. We explored two
different scenarios with regards to throughput and capacity enhancement; an MS communicating
directly with a BS and an MS communicating with a BS through RS.
In terms of link performance, relay link performed appreciably better than direct link in all
performance measures studied (i.e. BER, spectral efficiency and link capacity). This is attributed
to an increase in bits energy to noise ratio that relays offer which is essential for high data rate
transmission. In terms of multipath fading, it was concluded that fading channels significantly had
lower performance compared to non-fading AWGN channel. Mobile WiMAX is more affected by
the channel because of Doppler shift. To study different speeds, we used speeds of 30, 60 and 90
km/h and noted that as the speed increases, the performance of the system decreases.

50

International Journal of Wireless & Mobile Networks (IJWMN) Vol. 8, No. 3, June 2016

Latencies on both links were not considered in this work due to a negligible delay difference in
the single cell scenario.
In conclusion, the impact of deploying relays was thoroughly evaluated through link level
analysis. The analytical framework was presented highlighting on improved performance on
BER, Spectral Efficiency and Capacity as a benefit of using relay enhanced network. The results
also show how factors such as speed and fading affect performance. The simulation results will
serve as a baseline for future work on coverage and cell edge problems and their
countermeasures. Future work may incorporate real world involvement where drive test and
optimization tools can be used to achieve a more practical result.

REFERENCES
[1] P. Ramjee, and F. J. Velez. "OFDMA WiMAX Physical Layer." InWiMAX Networks, pp. 63-135.
Springer Netherlands, 2010.
[2] V. Ruby, and P. Garg. "Interpretation of IEEE 802.16 e (Wimax). "GJCST-E: Network, Web &
Security 13, no. 10 (2013).
[3] H.-C. Lu, W. Liao, and F.Y.-S Lin. "Relay Station Placement Strategy in IEEE 802.16j WiMAX
Networks," Communications, IEEE Transactions on , vol.59, no.1, pp.151,158, January 2011
[4] K. D. Satish, and N. Nagarajan. "A Survey on Technical Issues in IEEE 802.16 j Mobile Multi-hop
Relay Networks." SmartCR 1, no. 1 (2011): 12-33.
[5] IEEE Standard for Local and metropolitan area networks Part 16: Air Interface for Broadband
Wireless Access Systems Amendment 1: Multihop Relay Specification, IEEE Std 802.16j-2009
(Amendment to IEEE Std 802.16-2009) , vol., no., pp.1,290, June 12 2009
[6] J. Y. Chang. "An accelerating handover process scheme for IEEE 802.16 j multi-hop relay
networks." EURASIP Journal on Wireless Communications and Networking 2013, no. 1 (2013): 112.
[7] V. Genc, S. Murphy, Y. Yu, and J. Murphy. IEEE 802.16J relay- based wireless access networks : an
overview. Wireless Communications, IEEE, Vol.15, 2008.
[8] W. S. Peters, and W. H. Robert The future of WiMAX : Multihop relaying with IEEE 802.16j.
Communications Magazine, IEEE, January 2009.
[9] R. Pabst, B. Walke, D. Schultz, P. Herhold, H. Yanikomeroglu, S. Mukherjee, H. Viswanathan, M.
Lott, W. Zirwas, M. Dohler, H. Aghvami, D. Falconer, and G. Fettweis, Relay-based deployment
concepts for wireless and mobile broadband radio," IEEE Communications Magazine, vol. 42, pp. 8089, September 2004.
[10] N. Esseling, B. Walke, and R. Pabst, Fixed relays for next generation wireless systems," Emerging
Location Aware Broadband Wireless Ad Hoc Networks, Springer Science and Business Media, pp.
71-91, 2005.
[11] H. Li, M. Lott, M. Weckerle, W. Zirwas, and E. Schulz, Multihop communications in future mobile
radio networks," in The 13th IEEE International Symposium on Personal, Indoor and Mobile Radio
Communications (PIMRC), September 2002.
[12] H. Wei and R. Gitlin, Two-hop-relay architecture for next-generation WWAN/WLAN integration,"
IEEE Wireless Communications, vol. 11, pp. 24-30, April 2004.
[13] X. J. Li and P. Chong, A dynamic channel assignment scheme for TDMA-based multihop cellular
networks," IEEE Transactions on Wireless Communications, vol. 7, pp. 1999-2003, June 2008.
[14] J. De Bruyne, W. Joseph, L. Verlook, and L. Martens. Measurements and evaluation of the network
performance of a fixed WiMAX system in a suburban environment. IEEE International Symposium
on Wireless Communication Systems (ISWCS), 2008.
[15] K. Haneda, V. M. Kolmonen, T. Riihonen, R. Wichman, P. Vainikainen, and J. I. Takada.
Evaluation of relay transmission in outdoor-to-indoor propagation channels. In European
Cooperation in Science and Technology Workshop (COST) (Vol. 2100). 2008
[16] R. Patil. On throughput performance of decode and forward cooperative relaying with packet
combining and ARQ. International Journal of Computer Networks and Communications (IJCNC),
4(3), 57-67. (2012).
[17] K. Park, H. S. Ryu, C. G. Kang, D. Chang, S. Song, J. Ahn, and J. Ihm. The performance of relayenhanced cellular OFDMA-TDD network for mobile broadband wireless services. EURASIP
Journal on Wireless Communications and Networking, 2009.
51

International Journal of Wireless & Mobile Networks (IJWMN) Vol. 8, No. 3, June 2016

[18] R. Hu, C. Hu, J. Jiang, X. Xie, and L. Song. Full-duplex mode in amplify-and-forward relay
channels: outage probability and ergodic capacity. International Journal of Antennas and
Propagation, 2014.
[19] P. Gong, P. Xue, D. Park, and D. K. Kim. Optimum power allocation in a nonorthogonal amplifyand-forward relay-assisted network. Vehicular Technology, IEEE Transactions on, 60(3), 890-900.
[20] A. Azari, J. S. Harsini, and F. Lahouti. Power Allocation in Multi-hop OFDM Transmission Systems
with Amplify-and-Forward Relaying: A Unified Approach. arXiv preprint arXiv:1508.01216. 2015
[21] A. So and B. Liang, Enhancing WLAN capacity by strategic placement of Tetherless relay points,"
IEEE Transactions on Mobile Computing, vol. 6, pp. 522-535, May 2007.
[22] A. So and B. Liang, Optimal placement of relay infrastructure in heterogeneous wireless mesh
networks by Bender's decomposition," in International Conference on Quality of Service in
Heterogeneous Wired/Wireless Networks (QShine), 2006.
[23] L. C. Wang, W.-S. Su, J.-H. Huang, A. Chen, and C.-J. Chang, Optimal relay location in multi-hop
cellular systems," IEEE Wireless Communications and Networking Conference (WCNC), April 2008.
[24] E. Amaldi, A. Capone, and F. Malucelli, Planning UMTS base station location: optimization models
with power control and algorithms," IEEE Transactions on Wireless Communications, vol. 2, pp. 939952, September 2003.
[25] K. Cho, W. Lee, D. Yoon, H. S. Jin, K. Kim, and S. K. Park. "Frequency Allocation Scheme for
Interference Avoidance in Cellular System with Fixed Relay," Digital Telecommunications, 2008.
ICDT '08. The Third International Conference on , vol., no., pp.130,134, June 29 2008-July 5 2008
[26] D. Niyato, E. Hossain, D. I. Kim, and Z. Han. Joint Optimization of Placement and Bandwidth
Reservation for Relays in IEEE 802.16j Mobile Multihop Networks. IEEE International Conference
on Communications (ICC), June 2009.
[27] B. Lin, P. Han Ho, Xie, L. Liang, and X. Shen. Optimal relay station placement in IEEE 802.16j
networks. IEEE International Conference on Wireless Communications and Mobile Computing
(IWCMC), 2007.
[28] B. Lin, P. Han Ho, Xie, L. Liang, and X. Shen. Relay Station Placement in IEEE 802.16j Dual-Relay
MMR Networks. IEEE International Conference on Communications (ICC), May 2008.
[29] H. C. Lu, W. Liao, and F.Y. Lin. Relay Station Placement Strategy in IEEE 802.16j WiMAX
Networks. IEEE Transactionon Communications, Vol.59 :151-158, January 2011.
[30] L. Huang, M. Rong, H. Shi, and D. Yu, Comparison of two frequency reuse schemes in fixed relay
system," in International Conference on Wireless Communications, Networking and Mobile
Computing (WiCOM), September 2005.
[31] V. Roman, Frequency reuse and system deployment in local multipoint distribution service," IEEE
Personal Communications, vol. 6, pp. 20-27, December 1999.
[32] H.-M. Zimmermann, J. Eberspacher, and M. Lott, Frequency reuse for cellular multi-hop networks,"
in IEEE 62nd Vehicular Technology Conference (VTC), September 2005.
[33] S.-S. Wang, H.-C. Yin, and S.-T. Sheu, Symmetric path selection in IEEE 802.16 multi-hop relay
networks with error prone links," IEEE 4th International Conference on Circuits and Systems for
Communications (ICCSC), May 2008.
[34] S. Ann, K. G. Lee, and H. S. Kim. A Path Selection Method in IEEE 802.16j Mobile Multi-hop
Relay Networks. 2008 International Symposium on a World of Wireless, Mobile and Multimedia
Networks, 2008.
[35] S. Ann, K. G. Lee, and H. S. Kim, A path selection method in IEEE 802.16j mobile multi-hop relay
networks," 2nd International Conference on Sensor Technologies and Applications
(SENSORCOMM), August 2008.
[36] V. Genc, S. Murphy, and J. Murphy. Performance Analysis of Transparent Relays in 802.16j MMR
Networks. Modeling and Optimization in Mobile, Ad Hoc, and Wireless Networks and Workshops.
WiOPT 6th International Symposium on, August 2008.
[37] L. Erwu, W. Dongyao, L. Jimin, S. Gang, and J. Shan, Performance evaluation of bandwidth
allocation in 802.16j mobile multi-hop relay networks," IEEE 65th Vehicular Technology Conference
(VTC), April 2007.
[38] J. Z. Zhifeng Tao, Koon Hoo Teo, Aggregation and concatenation in IEEE 802.16j mobile multihop
relay (MMR) networks," IEEE Mobile WiMAX Symposium, March 2007.
[39] S.-J. Lin, W. H. Sheen, I.-K. Fu, and C. chi Huang, Resource scheduling with directional antennas
for multi-hop relay networks in manhattan-like environment," IEEE Mobile WiMAX Symposium,
March 2007.
52

International Journal of Wireless & Mobile Networks (IJWMN) Vol. 8, No. 3, June 2016

[40] I.-K. Fu, W.-H. Sheen, and F.-C. Ren, Deployment and radio resource reuse in IEEE 802.16j multihop relay network in manhattan-like environment," 6th International Conference on Information,
Communications and Signal Processing (ICICS), December 2007.
[41] Y. Sun, Y. Bian, P. Strauch, and A. Nix, Study of radio resource sharing for future mobile WiMAX
applications with relays," IEEE Mobile WiMAX Symposium, March 2007.
[42] M. Salem, A. Adinoyi, M. Rahman, H. Yanikomeroglu, D. Falconer, Y. D. Kim, E. Kim, and Y.-C.
Cheong, An overview of radio resource management in relay-enhanced OFDMA-based networks,"
IEEE Communications Surveys and Tutorials, 2009.
[43] Z. Tao, A. Li, K. H. Teo, and J. Zhang. Frame Structure Design for IEEE 802.16j Mobile Multihop
Relay (MMR) Networks. Global Telecommunications Conference, IEEE GLOBECOM, December
2007.
[44] C. Hoymann, M. Dittrich, and S. Boebbels. Dimensioning Cellular Wimax Part I : Singlehop
Networks. European Wireless, Paris, France, April 2007.
[45] C. Hoymann, M. Dittrich, and S. Boebbels. Dimensioning Cellular Wimax Part II : Multihop
Networks. European Wireless, Paris, France, April 2007.
[46] B. Upase and M. Hunukumbure. Dimensioning and Cost Analysis of Multihop Relay-Enabled
WiMAX Networks. Futjisu Scientific and Technical Journal, July 2008.
[47] H. Zeng and C. Zhu, System-level modeling and performance evaluation of multi-hop 802.16j
systems," International Wireless Communications and Mobile Computing Conference (IWCMC),
August 2008.
[48] E. Visotsky, J. Bae, R. Peterson, R. Berryl, and M. Honig, On the uplink capacity of an 802.16j
system," IEEE Wireless Communications and Networking Conference (WCNC), April 2008.
[49] N. R. Van, and R. Prasad. OFDM for wireless multimedia communications. Artech House, Inc.,
2000.
[50] Q. Zhang, S. Kassam. Finite-state Markov model for Rayleigh fading channels, IEEE Trans.
Commun. 47(11): 16881692. 1999.
[51] V. Erceg, K. V. S. Hari. Channel models for fixed wireless applications, IEEE 802.16 Broadband
Wireless Access Working Group, Technical Report, 2001
[52] T. Zhang, L. Xiao, C. Feng, and L. Cuthbert. "System Level Performance of Multiuser Diversity in
Cooperative Relay Based OFDMA Networks,"Advances in Mesh Networks, 2009. MESH 2009.
Second International Conference on , vol., no., pp.85,89, 18-23 June 2009
[53] L. Xiao, T. Zhang, Y. Zhu, and L. Cuthbert. "Two-Hop Subchannel Scheduling and Power Allocation
for Fairness in OFDMA Relay Networks,"Wireless and Mobile Communications, 2009. ICWMC '09.
Fifth International Conference on , vol., no., pp.267,271, 23-29 Aug. 2009

53