paper3

978-1-4799-2439-4/13/$31.00 © 2013 12
Efficient Link Adaptation in OFDM Systems using a Hybrid Intelligent Technique
1,2
Atta-ur-Rahman 2,3
Ijaz Mansoor Qureshi 4
Muhammad Hammad Salam 1,2
Muhammad Tahir Naseem
1
Barani Institute of Information Technology, Rawalpindi, Pakistan
2
Institute of Signals, Systems and Soft-computing (ISSS), Islamabad, Pakistan
3
Department of Electrical Engineering, Air University, Islamabad, Pakistan
4
The University of Lahore, Islamabad Campus, Islamabad, Pakistan
ataurahman@biit.edu.pk imqureshi@mail.au.edu.pk hamad.salam@yahoo.com tahir.naseem@biit.edu.pk
Abstract— Adaptive communication for rate enhancement and
link survival is a recent area of interest. In adaptive
communication, transmission parameters like transmit power,
code rate and modulation scheme are chosen in an adaptive
manner to match the link conditions and quality of service
demand at that transmission interval. But this is a highly non
linear phenomenon and a constrained optimization problem.
In this paper, we have investigated Simulated Annealing
algorithm in conjunction with a fuzzy rule base system (SA-
FRBS) for adaptive coding, modulation and power in an
orthogonal frequency division multiplexing environment. The
cases of fixed and adaptive communication are compared and
results are shown through the simulations.
Keywords-component; Simulated Annealing; OFDM; FRBS;
BER; Adaptive Modulation and Coding
I. INTRODUCTION
Evolutionary, soft-computing and hybrid intelligent
techniques have gained attention of many complex systems
over past decade. Many soft-computing algorithms like
genetic algorithms (GA), particle swarm optimization (PSO)
and simulated annealing (SA) are proposed a number of
decades ago, but they are being reinvestigated due to the
availability of fast computing hardware.
Orthogonal Frequency Division Multiplexing (OFDM) is
the prominent candidates for many 3rd
Generation (3G) and
4th
Generation (4G) Communication Systems. In this
technique a single very high data rate stream is divided into
several low data rate streams using Inverse Fast Fourier
Transform (IFFT). Then these streams are modulated over
different orthogonal subcarriers. This is to divide one large
frequency selective channel into a number of frequency non-
selective sub-channels. Moreover, addition of appropriate
cyclic prefix (CP) and interleaver makes the system almost
inter-symbol-interference (ISI) free.
In OFDM system, every subchannel experiences a different
channel condition that the use of same set of transmission
parameters would not be equally suitable for all subchannels.
For example, in the case of transmit power; some
subchannels need more or less power for the sake of
sustenance of the link according to their individual channel
state information (CSI) and/or to fulfill the data rate demand
at that time. This situation demands adaptive radio resource
allocations.
A Genetic Algorithm (GA) based adaptive resource
allocation scheme was proposed by Reddy [1], to increase
the user data rate where Water-filling principle was used as a
fitness function. Moreover, it was shown that chromosome
length helps to achieve optimum power requirement. A
subchannel allocation based on bidding model and auction
algorithm proposed by [2], where throughput was sustained
but user data rates were compromised. Moreover, auction
algorithm use to allocate subchannel to the appropriate most
user who require that subchannel. A novel efficient resource
allocation algorithm for multiuser OFDM system using a
joint allocation method and root finding algorithm to achieve
good performance even with low signal to noise ratio (SNR)
was proposed by [3]. Another interesting paper with adaptive
resource allocation based on modified GA and particle
swarm optimization (PSO) for multiuser OFDM system was
proposed by [4]. GA has been modified by using a fractional
generation gap. It converges faster than the original one and
it was found that PSO performs better than GA.
An approach akin to the previous one, ant colony
optimization (ACO) evolutionary technique for subcarrier
allocation in OFDMA-based wireless system was proposed
by [5]. Technique was capable of finding one optimal
solution in a short period of time. Adaptive subcarrier and
power allocation with fairness for multi-user space-time
block-coded OFDM system was investigated in contrast to
Greedy algorithm as well as water-filling principle [6].
An optimization problem for power constraints and use GA
to maximize the sum capacity of OFDM system with the
total power constraint was investigated in [7]. Also it was
shown that GA is better than conventional methods.
A scheme for resource allocation in downlink MIMO-
OFDMA with proportional fairness where dominant Eigen
channels obtained from MIMO state matrix are used to
formulate the scheme with low complexity in [8], scheme
provides much better capacity gain than static allocation
method. A PSO based Adaptive multicarrier cooperative
communication technique which utilizes the subcarrier in
deep fade using a relay node in order to improve the
bandwidth efficiency [9] where centralized and distributed
versions of PSO were investigated.
A low complexity subcarrier and power allocation
technique based upon GA to maximize the sum of user data
rates in MIMO-OFDMA system was proposed in [10].
Another GA based efficient real-time subcarrier and bit
allocation for multiuser OFDM transmission technique was
proposed in which overall transmit power was minimized
under user constraint [11].
A subcarrier-chunk based technique in which resource
allocation problem for the downlink of Orthogonal

132013 13th International Conference on Hybrid Intelligent Systems (HIS)
Frequency Division Multiple Access (OFDMA) wireless
systems was proposed in [12]. The scheme dramatically
reduces the complexity and fairness among users’ data rates
is very satisfactory despite the loss with respect to the
unconstrained case where the only target is the maximization
of the sum data rate.
In [13] the authors proposed a Fuzzy Rule Based System
(FRBS) for adaptive coding and modulation in OFDM
systems where quadrature amplitude modulation (QAM) and
convolutional codes were used as forward error correction
(FEC) codes and modulation schemes, respectively. In [14],
same authors proposed FRBS for adaptive coding and
modulation where Product codes were used as FEC. In both
of these papers, power was kept constant while code rate and
modulation was adaptive. In [15], same authors used GA and
Water-filling principle in conjunction with FRBS for
adaptive coding, modulation and power in OFDM systems,
where GA was used to adapt the power. It was found that
GA assisted adaptive power case performs better than water-
filling principle in terms of channel capacity. In [16], authors
investigated differential evolution (DE) algorithm with
FRBS for adaptive coding, modulation and power. A rate
enhancement scheme for OFDM based HYPERLAN was
proposed in [17] in which GA was used for adaptation.
In this paper simulated annealing algorithm with Fuzzy
Rule Base System (SA-FRBS) is proposed for adaptive
coding, modulation and power in OFDM system for rate
enhancement according to the individual subchannel CSI.
The remainder of this paper is organized as follows. In
section 2, system model is introduced. Performance of coded
modulation is presented in section 3. Section 4 formulates a
constrained optimization problem. In section 5 a brief
introduction to FRBS is given. Section 6 contains a brief
introduction of SA; Section 7 contains the performance
comparison of the scheme, while section 8 concludes the
paper.
II. SYSTEM MODEL
The system model considered is OFDM equivalent baseband
model with N number of subcarriers. It is assumed that
complete channel state information (CSI) is known at
receiver. The frequency domain representation of system is
given by;
. . ; k 1,2,......,k k k k kr h p x z N= + = (1)
where kr , kh , kp , kx and kz denote received signal,
channel coefficient, transmit amplitude, transmit symbol and
the Gaussian noise of subcarrier 1,2,......,k N= ,
respectively. The overall transmit power of the system is
1
N
total kk
P p=
= ∑ and the noise distribution is complex
Gaussian with zero mean and unit variance.
It is assumed that signal transmitted on the kth subcarrier
is propagated over Rayleigh flat fade channel and each
subcarrier faces a different amount of fading independent of
each other. This can be given mathematically as;
; 1,2,......,kj
k kh e k Nθ
α= = (2)
where kα is Rayleigh distributed random variable of kth
subcarrier, and the phase kθ is uniformly distributed over
[ ]0,2π . The proposed adaptation model is given in Fig-1.
Figure 1. Brief diagram of proposed System
III. CODED MODULATION
In this section performance of standard modulation and
codes being used in IEEE 802.11n/g/b are analyzed in terms
of bit error rate (BER) and SNR. For experimentation the
sequence of operations is carried out in same way as given in
fig-2. Following is the detail of each component.
A. Coding Scheme
The codes used in adaptive coding and modulation are
non-recursive convolutional codes with code rates taken
from the set C with constraint length 7. Set C is given
below;
C {1/ 4,1/3,1/ 2,2/3,3/ 4}= (3)
B. Modulation Scheme
In this paper we have utilized Quadrature Amplitude
Modulation (M-QAM) for adaptive coding and modulation,
with rectangular constellation. The modulation symbols are
taken from the following set. Set M is given by;
M = {2,4,8,16,32,64,128} (4)
C. Channel
Additive White Gaussian Noise (AWGN) channel is
assumed for simulations. This channel is proven to be a
good representative of channel condition at OFDM
subcarrier.
Figure 2. Brief diagram of simulations
Bit
loading
FEC
Encoder
QAM
Modulator
AWGN
Channel
Bit
Receiving
FEC
Decoder
QAM
Demodulator
OFDM PHY
Transmitter
OFDM
Channel
PHY layer
Receiver
Link Adaptation using
Simulated Annealing and Fuzzy
Rule Based System (SA-FRBS)
Quality of Service
(QoS) Demand/
Subcarrier
Feedback Channel
Sub-channel Estimates
New
Modulation
Code rate
Power

14 2013 13th International Conference on Hybrid Intelligent Systems (HIS)
The total number of MCPs can be given by;
x {( , ); , }i j i jP C M c m c C m M= = ∀ ∈ ∀ ∈ (5)
Then graph for each MCP is obtained and some of these
graphs are depicted in the fig-3 and fig-4 according to the
sequence of operations shown in fig-2.
Figure 3. BER comparison of different QAM using rate 1/4 code
Figure 4. BER comparison of different QAM using rate 1/2 code
IV. RATE OPTIMIZATION
In order to maximize the data rate for the overall OFDM
system, following constrained optimization problem is
considered.
1
1
1
max
s.t,
(6)k
N
Total k
k
k QoS
N
Total k T
k
R r
N
BER BER
and
P p P
=
=
=
≤
= <
∑
∑
where 2(log ( ))k k kr M R= is the bit rate of kth subcarrier
which is product of code rate and bits/symbol. TP is the total
transmit power and kQoSBER is target BER that depends
upon a specific quality of service (QoS) request or
application requirement over ith subcarrier, while N
represents the total number of subcarriers in OFDM system.
V. FUZZY RULE BASE SYSTEM
In this section FRBS is designed for optimum selection MCP
per subcarrier based upon received SNR and QoS. The steps
involved in creation of FRBS are described below.
A. Data Acquisition
From the results obtained in section-III, those code-
modulation pairs that fulfill different BER demands
depending upon different quality of services i.e.
5 4 3 2
10 ,10 ,10 ,10TBER − − − −
= etc are obtained. This is
obtained by drawing straight horizontal lines on the
graphs shown in figures 2-5, on certain BER values.
Then the points of intersection of these lines and the
curves (representing a code and a modulation) are noted
and according SNR value is noted. This information can
be expressed as “for a given SNR and specific QoS
which modulation code pair can be used”.
B. Rule Formulation
Rules for every pair are obtained by the appropriate
fuzzy set used. That is by putting complete pair in
input/output set and a rule generated for each pair.
C. Elimination of Conflicting Rule
The rules having same IF part but different THEN
parts are known as conflicting rules. This appears when
more than one modulation code pair (MCP) are available
for given specification. For instance, there is a rule
whose THEN part contains three different MCP namely,
[8, 1/2], [16, 2/3] and [16, 3/4]. Now [16, 3/4] is best
among the rest since its throughput is 4x3/4=3 while
others have 3x1/2=1.5 and 4x2/3=2.67 respectively.
Similarly, sometime there could be two different
pairs with same throughput like [2, 1/2] and [4, 1/4] both
have same throughput that is 1x1/2=0.5, then [2, 1/2] is
chosen since it exhibits less modulation/demodulation
and coding/decoding cost.
D. Completion of Lookup Table
Since in lookup table scheme we may not have
complete number of IO pairs, then those parts are filled
by heuristic or expert knowledge. For example, a
modulation code pair is suggested by rule for a certain
SNR and QoS. Then that rule can also be used for
slightly above SNR and poor QoS. For instance,
[128,3/4] is suggested for 25dB SNR and BER 3
10−
,
then this pair can be used for 26-30dB SNR and 2
10−
BER cases as well. Since if a modulation code pair
performs for lower SNR, then it can easily sustain in
higher SNR situations. Similarly, if a MCP performs for
a good QoS then it can sustain for poor QoS demands.

E. Fuzzy Rule Base Creation
Using the Lookup table in above phase, Fuzzy Rule
Base is created using Fuzzy Logic Toolbox in
MATLAB. Further details are given in next section.
Table look-up scheme for design of this fuzzy rule base
system is used.
The input-output pairs for design of FRBS are of the form;
1 2( , ; ); 1,2,3.......s s s
x x y s S= (7)
where 1
s
x represents received SNR, 2
s
x represents required
BER (QoS) and s
y represents the output MCP suggested by
FRBS, so the rule format can be given as below;
{IF ( 1x is L1 and 2x is Q7) THEN y is P2}
Following is the brief description of different components of
fuzzy rule based system used. Design of the FRBS is carried
out in MATLAB 7.0 standard Fuzzy System Toolbox.
• Fuzzy Sets
Sufficient numbers of fuzzy sets are used to cover
the input output spaces. There are two input variables
namely received SNR and quality of service (QoS)
demand while there is one output variable for
modulation code pair MCP. There are thirty-one, sixteen
and twenty-five fuzzy sets are used for the two input and
one output variable respectively.
• Fuzzifier
Standard triangular fuzzifier is used with AND as
MIN and OR as MAX. This is because the triangular
fuzzifier is suitable for real time applications due to its
simplicity.
• Rule Base
Rule base contains rules against all the IO pairs. As
there are thirty-one sets (L0 to L30) for first input variable
named SNR and about sixteen sets (Q1 to Q16) for input
variable MLBER. Hence there are 496 rules in rule base
and rule base is complete.
• Inference Engine
Standard Mamdani Inference Engine (MIE) is used
that will infer which input pair will be mapped on to
which output point.
• De-Fuzzifier
Standard Center Average Defuzzifier (CAD) is used
for defuzzification due to its reasonable simplicity and
efficiency.
Fig-5 shows the impact on throughput for different values of
SNR and QoS demand after incorporating the constraint.
Throughput reaches to its maximum for quality of service
value two (BER=10e-2) and SNR value 30dB, similarly for
quality of service value five (BER=10e-5) and poor SNR
values, throughput is reduced.
Figure 5. Rule surface
VI. SIMULATED ANNEALING ALGORIHTM
Simulated annealing (SA) is a global optimization algorithm
that combines statistical mechanics and combinatorial
optimization. It was developed by Kirkpatrick et al. in
1983[18]. It is very famous for finding global optimum in
very large search spaces. Its name originates from the
metallurgy process annealing, a technique involving heating
and controlled cooling of a material to increase the size of
its crystals and reduce their defects. The heat causes the
atoms to become unstuck from their initial positions (a local
minimum of the internal energy) and wander randomly
through states of higher energy; the slow cooling gives them
more chances of finding configurations with lower internal
energy than the initial one. SA has a large number of
applications in bio-informatics, engineering and other
disciplines. In this technique we have to choose next state
based upon fitness criteria.
In our proposal the fitness function is given in fig-6. In
order to find the optimum power vector, the basic flat power
vector (initial guess) is passed through the OFDM system.
Once the state is known, using the SA we find the optimum
vector that gives us the highest throughput. The calculated
throughput is based upon the modulation code pairs (MCPs)
obtained by FRBS. So in our case, it is apparent that FRBS
is used as a fitness function in SA.
Figure 6. Fitness Block
TransmitPowerVector(P)
2
1α
2
2α
2
Nα
FuzzyRuleBaseSystem(FRBS)
Throughput
(MCP) 1
(MCP) 2
(MCP)N
1
1 N
i
i
r
N =
∑
Quality of Service Vector Q

16 2013 13th International Conference on Hybrid Intelligent Systems (HIS)
VII. RESULTS
In this section proposed scheme is compared with other
schemes. Table-1 contains the simulation parameters.
TABLE-1 Simulation Parameters
Sr. Parameter Value
1 Number of Subcarriers N 1024
2 Fitness Function for SA Fuzzy Rule Base System Fig-
6
3 SA iterations 30
4 Channel considered for
simulation
IEEE 802.11n
indoor channel (WIFI)
5 Channel Coefficients range [0.1-0.4]
6 Quality of Service (QoS) 10e-2,10e-3,10e-4 and 10e-5
7 Adaptive Criterion SA-FRBS
8 Parameters being adapted Code rate, Modulation and
power
Fig-7 to Fig-10 show the comparison of simulated annealing
algorithm and Fuzzy Rule Base System (SA-FRBS) assisted
Adaptive Coding, Modulation and Power (ACMP) scheme
for different values of target BER. In fig-7 target BER was
fixed at 10e-2, the proposed scheme approaches to
5.5bits/s/Hz for an SNR value of 25dB, while in contrast the
adaptive coding and modulation scheme with fixed power
per subcarrier it reaches till 5bits/s/Hz at 30dB SNR. For
obtaining 5bits/s/Hz with adaptive power, 22.5dB is
required, that states a 7dB gain over fixed power case.
In fig-8 same kind of difference can be noted where
target BER is 10e-3. However, here 27dB is needed to
achieve 5.5bits/s/Hz for adaptive power case, so 2dB more
power is required because the required BER is 10e-3. In fig-
9, at 25dB, adaptive power case provides a throughput of
4.5bits/s/Hz while at the same SNR value the fixed power
case gives 3bits/s/Hz. So rate enhancement with adaptive
power is 1.5bits/s/Hz with a target BER of 10e-4. In fig-10,
fixed and adaptive power is investigated for target BER of
10e-5, at 30dB the adaptive power is 0.2bits/s/Hz better than
the fixed case. However, in order to achieve 2bits/s/Hz still
the gain is 7dBs in adaptive power over the fixed power
scheme. At 15dB, however result is very dramatic, that is
both fixed and adaptive power schemes are performing
identically.
Figure 7. Comparison of proposed scheme with QoS=10e-2 per subcarrier
Figure 8. Comparison of proposed scheme with QoS=10e-3 per subcarrier
Figure 9. Comparison of proposed schemes with QoS=10e-4 per subcarrier
Figure 10. Comparison of proposed schemes with QoS=10e-4 per subcarrier

In fig-11, the proposed scheme is compared for various
target BERs. For a relaxed target BER, that is 10e-2 and 10e-
3 the throughput reaches to its maximum and with a stringent
BER that is 10e-5, the adaptive power still plays a vital role
and makes the transmission survive.
Figure 11. Comparison of proposed scheme for different target BER
The above simulations are carried out using optimization
toolbox in MATLAB 7.0.
VIII. CONCLUSIONS
In this paper simulated annealing and fuzzy rule based
system (SA-FRBS) assisted adaptive coding, modulation
and power scheme is proposed for rate enhancement in
OFDM Systems. Performance of the proposed scheme has
been investigated over IEEE 802.11n (WIFI) environment
for IEEE standard indoor channel. Simulation results show
the viability of the proposed scheme and its significance in
rate enhancement compared to its fixed power variant.
Performance is measured for different quality of service
demands (target BER) per subcarrier in OFDM system. In
most of the cases 7dB gain is achieved in case of adaptive
power compared to the fixed power case.
REFERENCES
[1] Reddy, Y.B.; Gajendar, N.; Taylor, Portia; Madden, Damian,”
Computationally Efficient Resource Allocation in OFDM Systems:
Genetic Algorithm Approach “, Dept of Math & Comput. Sci.,
Grambling State Univ., LA, 2007, pp 36 – 41.
[2] Jinyoung Oh, Sang-wook Han,Youngnam Han,“Efficient and fair
subchannel allocation based on auction algorithm”, 2008, pp: 1 – 5.
[3] K. Gunaseelan, R. Venkateswari, A. Kandaswamy, ”A novel efficient
resource allocation algorithm for multiuser OFDM systems”, 2008,
pp: 201-206.
[4] Ahmed, I.; Majumder, S.P.,“Adaptive resource allocation based on
modified Genetic Algorithm and Particle Swarm Optimization for
multiuser OFDM systems”, Dept. of Electr. & Electron. Eng.,
Bangladesh Univ. of Eng. & Technol., Dhaka, 2008, pp: 211 – 216.
[5] Ioannis Chatzifotis, Kostas Tsagkaris, Panagiotis Demestichas,“Ant
colony optimization for subcarrier allocation in OFDMA-based
wireless system”, 2009.
[6] Jian xu, Jong-soo seo,“Adaptive subcarrier and power allocation with
fairness for multi-user space-time block-coded OFDM system”, 2009,
pp:164-177.
[7] Bo Liu,Mingyan Jiang, Dongfeng Yuan,“Adaptive Resource
Allocation in Multiuser OFDM System Based on Genetic
Algorithm“, Sch. of Inf. Sci. & Eng., Shandong Univ., Jinan , 2009,
pp: 270 – 273.
[8] Fuwa,Y. Okamoto, E.Iwanami, Y.,“Resource allocation scheme with
proportional fairness for multi-user downlink MIMO-OFDMA
systems”,Dept. of Comput. Sci. & Eng., Nagoya Inst. of Technol.,
Nagoya, Japan, 2009, pp: 588 – 593.
[9] Chilukuri Kalyana Chakravarthy, Prasad retty,”Particle swarm
optimization based approach for resource allocation and scheduling in
OFDMA systems”, 2010, pp:467-471.
[10] Nitin sharma. K.R.Anupama,”A novel genetic algorithm for adaptive
resource allocation in MIMO OFDM systems with proportional rate
constraint”, Wireless Personal Communication SpringerLink, vol. 61,
No. 1, pp. 113-128, 2010.
[11] Elhem chriaa, Mohamed quzineb and brunilde sanso,” Genetic
algorithm for efficient real-time subcarrier and bit allocation for
multiuser OFDM transmission “, 2011.
[12] Papoutsis, V.D., Kotsopoulos, S.A.,“Chunk-Based Resource
Allocation in Distributed MISO-OFDMA Systems with Fairness
Guarantee “,Dept. of Electr. & Comput. Eng., Univ. of Patras, Rio,
Greece 2011,pp: 377 – 379.
[13] Atta-ur-Rahman, Qureshi I.M., Malik A.N.,“A Fuzzy Rule Base
Assisted Adaptive Coding and Modulation Scheme for OFDM
Systems”, J. Basic Appl. Sci. Res. Vol. 2(5), pp. 4843-4853, 2012.
[14] Atta-ur-Rahman, Qureshi I.M. and Muzaffar M.Z. “Adaptive Coding
and Modulation for OFDM Systems using Product Codes and Fuzzy
Rule Base System”. International Journal of Computer Applications
(IJCA), Vol. 35(4), pp.41-48, December 2011.
[15] Atta-ur-Rahman, Qureshi I.M., Malik A.N.,“Adaptive Resource
Allocation in OFDM Systems using GA and Fuzzy Rule Base
System”, World Applied Sciences Journal, Vol. 18(6), pp. 836-844,
2012.
[16] Atta-ur-Rahman, Qureshi I.M., Malik A.N., Naseem M.T., “Dynamic
Resource allocation for OFDM Systems using Differential Evolution
and Fuzzy Rule Base System”, Journal of Intelligent & Fuzzy
Systems (JIFS), DOI: 10.3233/IFS-130880, June, 2013.
[17] Atta-ur-Rahman, Qureshi I.M., Naseem M.T., Muzaffar M.Z., “A
GA-FRBS based Rate Enhancement Scheme for OFDM based
Hyperlans”. IEEE 10th
International Conference on Frontiers of
Information Technology (FIT’12), pp-153-158, December 17-19,
2012. Islamabad, Pakistan.
[18] S. Kirkpatrick, C. D. Gelatt, and M. P. Vecchi, “Optimizationby
simulated annealing,” Science, vol. 220, no. 4598, pp. 671–680, 1983.

paper3

More Related Content

What's hot (20)

Viewers also liked (9)

Similar to paper3 (20)

paper3