Performance Evaluation of Adaptive Array Antennas in Cognitive Relay Network

Circuits and Systems: An International Journal (CSIJ), Vol.1, No.4, October 2014.
15
PERFORMANCE EVALUATION OF ADAPTIVE ARRAY
ANTENNAS IN COGNITIVE RELAY NETWORK
Eman Sadik1
, Mona Shokair2
, Abd Elmaged Sharshar3
and Said Elhalafawy4
1,2,3,4
Department of Electronics and Electrical Communication, Faculty of Electronic
Engineering, El- Menoufia University, Menouf, Egypt.
ABSTRACT
Adaptive Array Antennas (AAAs) are expected to play a key role in meeting the demands of the wireless
communication systems of the future. AAAs in cognitive relay network is proposed to reduce the symbol
error rate and improve the system performance. Many algorithms such as wiener solution and least mean
square (LMS) will be explained to show how AAAs in cognitive relay network achieves this object. AAAs
at different locations will be investigated under AWGN and Rayleigh fading channel. Moreover,
enhancement the system performance by showing the effect of increasing the number of AAAs element at
the relay node, increasing the source gain and decreasing the relay gain. In addition, increasing the rate
adaptation and number of iterations in LMS algorithm has significant improvement in the system.
KEYWORDS
Adaptive Array Antennas, cognitive relay network and least mean square algorithm
1. INTRODUCTION
Cognitive radio is used to designate intelligent wireless communication system that is able to
adapt changes occurring in the surrounding environment [1]. It allows Secondary User (SU)
network to coexist with Primary User (PU) network through spectrum sharing, provided that the
secondary spectrum access will not affect the PUs performance [2], [3].
To enhance cognitive radio performance, the cooperative communications that it’s used in [4],
where one or more nodes help the communication between source and destination by acting as
relays, achieve spatial diversity even in a network composed of a single antenna device. In [5],
three protocols of cooperative communications are presented; Amplify and Forward (AF),
Decode and Forward (DF) and Adaptive Relay Protocol (ARP). In the AF protocol, the relay
amplifies the received signal and forwards it to the destination. In the DF protocol, the relay tries
to decode the source message and then re-encodes and forwards it to the destination. ARP is
used to overcome the disadvantage of these two protocols. In our system, AF protocol is used as
its low complexity than the other mentioned.
Another critical issue in cognitive radio is to improve the performance of the system by using
AAAs [6-8]. AAAs is one of the key technologies that are expected to dramatically improve
future wireless communication systems because it has the potential to expand coverage, increase
capacity and improve signal quality. AAAs algorithms, which use reference signal (desired
signal) will be explained in this paper. Other AAAs algorithms, which doesn’t use reference
signal are not considered here. Types of AAAs algorithms used with desired signal are wiener
solution, method of steepest descent, Least Mean Squares (LMS), Normalized Least Mean

16
Squares (NLMS) and Recursive Least Square (RLS). Wiener solution and LMS algorithms will
be used in this paper. In these algorithms, desired signal is sent during training period. By using
the received desired signal at AAAs, optimum weights can be computed. After the training
period, data is sent and AAAs use the weights computed previously to process the received
signal. AAAs in cognitive relay network will be analysis in this paper to give more improvement
in the system performance.
The reminder of this paper is organized as follows: Section II describes the system model for
the analysis of AAAs in cognitive relay network including wiener solution and LMS algorithms.
Simulation results are illustrated in Section III. Finally, conclusions are made in Section V.
2. SYSTEM MODEL
Figure 1. Adaptive array antenna in cognitive relay network.
AAAs in cognitive relay network at the source, relay and destination is illustrated in Figure 1,
where S, R and D denotes the source, relay and destination respectively. The relay node is
randomly located between the source and the destination to provide most enhancements for the
link. The common channels of S and D will be reserved for the usage of direct link. The
channels used by R for receiving and transmitting are called indirect channels. They are used for
amplifying the received data, and then retransmit it to the destination node.
An antenna array consists of N identical antenna elements arranged in a particular geometry,
where the geometry of the array determines the amount of coverage in a spatial region. AAAs
algorithms, which use the desired signal, will be used in this paper. In these algorithms, desired
signal is sent and by using the received desired signal at AAAs, optimum weights can be
computed. Then, data is sent and AAAs uses the previously computed weights to process the
received signal.

17
In the following AAAs algorithms are described which try to minimize the mean square error
between the desired signal d (k) and the array output signal y (k).
2.1. Wiener Solution
The error signal e (k) can be expressed as the difference between y (k) and d (k) at sampling
instant k as follows [8],
e ( k) = d(k) –y(k) (1)
Where d(k) is the desired signal and the mean squared error is defined by,
J(k) = E[|e(k)|
] = E[d(k) – y(k)

] (2)
The optimum weights control can be given as

(). = (k)
() (3)
Where = [()
(()∗
)
] is the autocorrelation matrix of the input signal (), =
[()(()
)∗
] is the cross correlation matrix between the complex conjugate of the input
signal and the desired signal.
The output y(k) of linear combination of data for M elements at time k is denoted as,
y(k) = ∑
!

!# ()!() (4)
In vector form the last equation can be written as,
Y(k) = X(k)
%() (5)
Where (. )
is the transpose operator.
2.2. LMS Algorithm
The block diagram of Least Mean Square algorithm is represented in Figure 2. The LMS is an
iterative beamforming algorithm that uses the estimate of the gradient vector from the available
data. The weight vector is updated in accordance with an algorithm that adapts to the incoming
data. The simplicity of LMS algorithm comes from the fact that it doesn’t require measurements
of the correlation functions nor matrix inversion. The updated weight is denoted by [9],
w(k+1) = w(k) + 2 () '∗
() (6)
where μ is defined as the rate of adaptation, controlled by the processing gain of the antenna
array (step size parameter) and e∗
(k) is the complex conjugate of error signal.

18
Figure 2. Block diagram representation of LMS algorithm.
3. SIMULATION RESULTS
In this section, the numerical results supporting the analysis of adaptive array antenna in
cognitive relay network are presented. Here, the cognitive relay network is demonstrated
consisting of a source, a destination and a cognitive relay node between them. The performance
of AAAs in cognitive relay network is illustrated using wiener solution and LMS algorithms
under AWGN and Rayleigh fading channel. Assume that, the number of transmitted symbols N
is equal to10+
. For simplicity, it is assumed that, the same transmitted power of source and
relay will be applied.
In the following the effect of many parameters will be studied as follows:
3.1. Effect of AAAs at different locations in cognitive relay network
Figure 3 compares Symbol Error Rate (SER) of AAAs at source, relay and destination node
using wiener solution algorithm. Numerical results are obtained by transmitted power of source
,- =
.

, transmitted power of relay ,/ =
.

. As shown in this figure, locating AAAs at the
source has better performance than locating AAAs at the relay and destination.

19
Figure 3. SER of AAAs at different locations in cognitive relay network.
As the AAAs at the relay node using wiener solution algorithm has the worst performance as
shown in Figure 3, some parameters are studied to enhance the performance of the system.
3.1.1. Increasing the source node gain
Figure 4 shows the numerical results of increasing the antenna gain at the source (0-) node
when AAAs at the relay node under using wiener solution algorithm. From this figure, it can be
concluded that significant improvement in performance is observed by increasing (0-) .
Figure 4. Impact of increasing the source gain when AAAs at the relay node.
3.1.2. Decreasing the relay node gain
The effect of decreasing the antenna gain at the relay (0/) is illustrated in Figure 5 when AAAs
at the relay node using wiener solution algorithm. It is shown in this figure that improving the
system performance is obtained by decreasing 0/ which explained by occurring interference on
the signal.
0 2 4 6 8 10 12 14 16 18 20
10
-5
10
-4
10
-3
10
-2
10
-1
10
0
SNR(dB)
SER
Wiener Solution Algorithm
AAAs at the Source
AAAs at the Relay
AAAs at the Destination
-10 -8 -6 -4 -2 0 2 4 6 8 10
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
SNR(dB)
SER
AAAs at the Relay Node using Winner Solution Algorithm
Gs=Gt/2
Gs=2Gt/3
Gs=3Gt/4

20
Figure 5. Effect of decreasing the relay gain when AAAs at the relay node.
3.1.3. Increasing AAAs elements
Figure 6 depicts the impact of increasing the number of antennas element of AAAs at the relay
node. We can conclude from this figure that, the performance of the system using AAAs with
three antennas element outperforms using AAAs with two antennas element.
Figure 6. SER at different AAAs elements.
3.1.3.1. Effect of different types of fading channels and noise
The simulative results of AAAs at the source node using wiener solution algorithm under
different fading channels are illustrated in Figure 7. For instance, at SNR =8dB, the SER is
approximately about 101
under AWGN, 10
under Rayleigh fading channel and 10
under
-10 -8 -6 -4 -2 0 2 4 6 8 10
0.25
0.3
0.35
0.4
0.45
0.5
0.55
0.6
0.65
0.7
0.75
SNR(dB)
SER
AAAs at the Relay Node using Winner Solution Algorithm
Gr=Gt/4
Gr=Gt/3
Gr=Gt/2
-10 -8 -6 -4 -2 0 2 4 6 8 10
0.35
0.4
0.45
0.5
0.55
0.6
0.65
0.7
0.75
SNR(dB)
SER
AAAs at the Relay Node using Wiener Solution Algorithm
AAAs with 2 antennas element
AAAs with 3 antennas element

21
AWGN plus Rayleigh fading channel. The numerical results indicate that AAAs at the source
under AWGN outperforms the other types.
Figure 7. AAAs at the source node under different types of fading channel using wiener
solution.
The effect of different fading channels of AAAs at the source using LMS algorithm is depicted
in figure 8. For instance, at SNR = 15dB, the SER is approximately 10
under AWGN and
102.3
under Rayleigh fading channel. The results show that AAAs at the source under AWGN
has better performance than under Rayleigh fading channel.
Figure 8. AAAs at the source node under different types of fading channel using LMS algorithm.
3.1.3.2. Effect of different step size parameter in LMS algorithm.
Figure 9 provides AAAs at the source using LMS algorithm with AWGN under different step
size parameters. As shown in this figure, by increasing the step size parameter, enhancing the
performance of the system.
0 2 4 6 8 10 12 14 16 18 20
10
-5
10
-4
10
-3
10
-2
10
-1
10
0
SNR(dB)
SER
AAAs at the Source using Wiener Solution Algorithm
Rayleigh Fading Channel
AWGN
Rayleigh Fading Channel+AWGN
10 15 20 25 30 35
10
-4
10
-3
10
-2
10
-1
10
0
SNR(dB)
SER
AAAs at the Source using LMS Algorithm
AWGN
Rayleigh Fading Channel

22
Figure 9. Compares different step size parameter in LMS algorithm.
3.1.3.3. Effect of different number of iterations in LMS algorithm.
AAAs at the source using LMS algorithm with AWGN under various number of iteration is
shown in Figure 10. From this figure, it’s found that; the performance of the system is improved
by increasing the number of iterations.
Figure 10. Shows different number of iteration in LMS algorithm.
3. CONCLUSIONS
In this paper, Adaptive array antennas in cognitive relay network are presented to enhance the
performance of the network. Wiener solution and least mean square algorithms are analyzed to
calculate the optimum weights. Putting AAAs at different locations such as at the source, relay
and destination are analytically derived. In addition, the system enhancement is investigated
under AWGN and Rayleigh fading channel is proposed. Moreover, increasing the step size
parameter and the number of iteration in LMS algorithm, improve the system performance. The
10 15 20 25 30 35 40
10
-4
10
-3
10
-2
10
-1
10
0
SNR(dB)
SER
AAAs at the Source using LMS with AWGN
mu=0.01
mu=0.02
mu=0.03
mu=0.04
0 2 4 6 8 10 12 14
10
-5
10
-4
10
-3
10
-2
10
-1
10
0
SNR(dB)
SER
AAAs at the Source using LMS with AWGN
No of Iteration=100
No of Iteration=1000

23
simulation results show that combining adaptive array antennas and cognitive relay network
present significant improvement in the performance of the system.
REFERENCES
[1] J. Mitola, (2000) “Cognitive Radio: An Integrated Agent Architecture for Software Defined Radio,”
Ph.D. dissertation, KTH Royal Inst. of Technol., Stockholm, Sweden.
[2] R. Manna,Y. Louie, Y. li and B. Vucetic, (2011) “Cooperative Spectrum Sharing in Cognitive Radio
Networks with Multiple Antennas”, IEEE Transactions on Signal Processing, vol. 59, no. 11.
[3] E. Sadik, M. Shokair, A. Sharshar and S. Elhalafawy, (2014) “Investigation of Multiple Antennas in
Cooperative Cognitive Relay Network ”, CiiT International Journal of Programmable Device
Circuits and Systems, vol. 6, no. 3, pp. 84-88.
[4] A. Nosratinia, T. E. Hunter, and A. Hedayat, (2004) “Cooperative Communication in Wireless
Networks”, IEEE Communications Magazine, vol. 42, no. 10, pp. 74-80.
[5] E. Sadik, M. Shokair and S. Elhalafawy, (2013) “Performance of Multiple Antennas Cognitive Relay
Networks ”, International Journal of Computer Applications, vol. 81, no. 1, pp. 27-32.
[6] M. Shokair and Y. Akaiwa, (2005) “A Feedback Type Adaptive Array Antenna with One Bit
Feedback Information and Adaptive Update Size in FDD System”, IEICE Transactions, vol. 88-B
(10), pp. 4074-4080.
[7] M. Shokair and Y. Akaiwa, (2006) “Performance of Feedback-type Adaptive Array Antenna in FDD
System with Rake Receiver”, IEICE Transactions on Communications, vol. 89-B, no. 1, pp. 539 -
544.
[8] M. Shokair and Y. Akaiwa, (2003) “The Performance of Feedback-type Adaptive Array Antenna in
FDD/CDMA System with Rake Receiver”, The 57th IEEE Semiannual Vehicular Technology
Conference, vol. 2.
[9] M. Yasin, P. Akhtar and Valiuddin, (2010) “Performance Analysis of LMS and NLMS Algorithms
for a Smart Antenna System”, International Journal of Computer Applications, vol.4, no. 9, pp. 0975
– 8887.
Authors
Eman Sadik was born in Menoufia, Egypt, in 1990. she received her B.Sc. degree in
electronics and electrical Communications engineering from the Faculty of Electronic
Engineering, Menoufia University, Egypt, in 2009.she is an Electrical Engineer at
Faculty of electronic engineer, Menoufia University. Her graduation project was about
open eNB with excellent degree. Her research interests include wireless mobile
communications and next generation networks.
Mona Shokair received the B.Sc., and M.Sc. degrees in electronics engineering from
Menoufia University, Menoufia,Egypt, in 1993, and 1997, respectively. She received the
Ph.D. degree from Kyushu University, Japan, in 2005. She received VTS chapter IEEE
award from Japan, in 2003. She published about 40 papers until 2011. She received the
Associated Professor degree in 2011. Presently, she is an Associated Professor at
Menoufia University. Her research interests include adaptive array antennas, CDMA
system, WIMAX system, OFDM system, and next generation networks.
Abdul-Mageed Sharshar received the B.Sc., and M.Sc. degrees in electronics
engineering from Menoufia University, Menoufia, Egypt, in 1978, and 1982,
respectively. He received the Ph.D. degree from England, London University, in 1991.
Presently, he is Lecturer at faculty of electronic engineering, Menoufia University. He
has published several scientific papers in national and international conferences and
journals. His current research areas of interest include Antennas and Microwaves.

24
Said Elhalafawy received the B.Sc., and M.Sc. degrees in electronics engineering from Menoufia
University,Menoufia, Egypt, in 1978, and 1984, respectively. He received the Ph.D.
degree from Plzen, Czech Republic, in 1990. Presently, he is the dean and Professor at
faculty of electronic engineering, Menoufia University. He has published several
scientific papers in national and international conferences and journals. His current
research areas of interest include image processing, speech processing, digital
communications and electromagnetic applications.

Performance Evaluation of Adaptive Array Antennas in Cognitive Relay Network

More Related Content

What's hot (17)

Similar to Performance Evaluation of Adaptive Array Antennas in Cognitive Relay Network (20)

Recently uploaded (20)

Performance Evaluation of Adaptive Array Antennas in Cognitive Relay Network