paper2

978-1-4799-3400-3/13/$31.00 © 2013 19
Adaptive Communication using Softcomputing Techniques
1,2,4
Atta-ur-Rahman 3
Ijaz Mansoor Qureshi 5
Muhammad Hammad Salam 4
Muhammad Zeeshan Muzaffar
1
Barani Institute of Information Technology, PMAS Arid Agriculture University Rawalpindi, Pakistan
2
Institute of Signals, Systems and Soft-computing (ISSS), Islamabad, Pakistan
3
Department of Electrical Engineering, Air University, Islamabad, Pakistan
4
School of Engineering & Applied Sciences (SEAS), ISRA University, Islamabad, Pakistan
5
School of Information Technology (SIT), The University of Lahore, Blue Area, Islamabad, Pakistan
ataurahman@biit.edu.pk imqureshi@mail.au.edu.pk hammad.salam@yahoo.com zeeshan.muzaffar@isra.edu.pk
Abstract— Adaptive communication has gained attention of
almost every recent communication system because of its rate
enhancement and context aware features. In this concept,
different transmission parameters like transmit power,
forward error correcting (FEC) code rate and modulation
scheme are adaptively chosen according to the channel state
information. Consequently, that set of transmission parameters
is chosen that maximizes the channel capacity as well as fulfills
the power and bit error rate constraints. Finding the optimum
value of the said parameters is a highly non-linear problem
with huge search space for solution. In this paper, we have
investigated Ant Colony Optimization (ACO) in conjunction
with a fuzzy rule base system (SA-FRBS) for adaptive coding,
modulation and power in an orthogonal frequency division
multiplexing environment. Proposed scheme is compared with
Simulated Annealing and FRBS (SA-FRBS) assisted adaptive
coding modulation and power scheme as well as with the fixed
power scheme. Superiority of proposed scheme is shown by the
simulations.
Keywords-component; ACO; Simulated Annealing; OFDM;
FRBS; BER; Adaptive Modulation and Coding
I. INTRODUCTION
Application of evolutionary and soft-computing algorithms
for solution of complex problems in engineering is an
emerging area of research nowadays. Nature and biologically
inspired algorithms like Ant Colony Optimization (ACO),
Bee Colony Optimization (BCO), Particle Swarm
Optimization and Genetic Algorithms are given a potential
importance in many areas of research presently.
Orthogonal Frequency Division Multiplexing (OFDM)
systems has gained attention of almost every present and
future communication systems due to promising data rates
and less vulnerability of frequency selectivity and inter-
symbol-interference (ISI) problem.
In OFDM systems each sub-channel has different type of
channel conditions so using same set of transmission
parameters like code rate, modulation scheme and transmit
power would not be equally suitable for all sub-channels. For
example, a sub-channel with good conditions can support a
high code rate and high modulation order. In contrast, a sub-
channel with poor conditions would need a relatively low
code rate and small modulation symbol. Similarly the need
of transmit power may vary from one sub-channel to
another. This situation demands adaptive parameter selection
rather than the fixed ones.
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

20 2013 International Conference of Soft Computing and Pattern Recognition (SoCPaR)
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
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] where GA was used.
In [18], authors proposed adaptive coding and modulation
scheme using machine learning framework that helps in
predicting the modulation and code rate based upon past
observations and CSI. In [19], authors proposed an outer
loop link adaptation for bit interleaved coded modulation
BICM-OFDM using adaptive kernel regression for learning.
In [20], a data driven approach to link adaptation is used
where machine learning classifier is used to select the
modulation and code rate.
In this paper ant colony optimization algorithm with Fuzzy
Rule Base System (ACO-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 ACO; Section 7 contains the performance
comparison of the scheme, while section 8 concludes the
paper.
II. SYSTEM MODEL
In this paper, standard OFDM channel is considered 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.
The proposed adaptation model is given in Fig-1.
Subchannel estimates provided by the PHY layer receiver
and quality of service demands per subcarriers is given to the
adaptation block, which in return provides the updated
parameter set. These new parameters are sent to the PHY
layer transmitter using a feedback channel.
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. The initial
set of experiments is carried over an additive white Gaussian
Noise (AWGN) channel. The sequence of simulations is
given in fig-2
A. Coding Scheme
The codes used as adapting coding parameters 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 adapting the modulation paramter,
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)
OFDM PHY
Transmitter
OFDM
Channel
PHY layer
Receiver
Link Adaptation using
evolutionary algorithms and
Fuzzy Rule Based System
Quality of Service
(QoS) Demand/
Subcarrier
Feedback Channel
Sub-channel Estimates
New
Modulation
Code rate
Power

212013 International Conference of Soft Computing and Pattern Recognition (SoCPaR)
Figure 2. Brief diagram of simulations
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 fig-3 using the sequence of operations
shown in fig-2.
Figure 3. 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 kr and number of modulation
bits/symbol 2(log ( ))kM . 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 is 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
Facts from the graphs are obtained by drawing straight
horizontal lines on the graphs on certain BER values.
Then the points of intersection of these lines and the
curves (a modulation code pair) are noted and according
SNR value is noted. This is shown in fig-4.
Figure 4. Obtaining facts from graphs
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.
D. Fuzzy Rule Base Creation
Using the Lookup table in above phase, Fuzzy Rule
Base is created using Fuzzy Logic Toolbox in
MATLAB. The rule format can be given as below;
{IF ( 1x is L1 and 2x is Q7) THEN y is P2}
Following are the components of FRBS.
• Fuzzy Sets
There are two input variables namely received SNR
and QoS. There is one output variable for modulation
code pair MCP. There are thirty-one sets (L0 to L30) for
first input variable named SNR and sixteen sets (Q1 to
Q16) for second input variable QoS; there are twenty-five
sets in output variable MCP.
Bit
loading
FEC
Encoder
QAM
Modulator
AWGN
Channel
Bit
Receiving
FEC
Decoder
QAM
Demodulator

• Fuzzifier& De-Fuzzifier
Standard triangular fuzzifier is used with AND as
MIN and OR as MAX. Standard Center Average
Defuzzifier (CAD) is used for defuzzification.
• Inference Engine
Standard Mamdani Inference Engine (MIE) is used
that will infer which input pair will be mapped on to
which output point.
VI. ADAPTIVE POWER APPROACHES
A. Ant Colony Optimization (ACO)
Ant colony optimization [21] is a nature inspired
algorithm that is based on the behavior of ant colony
searching for food in parallel. In ACO approach, several
artificial ants perform a sequence of operation
iteratively. In each iteration, several ants search in
parallel for good solution in the solution space. Ants that
hit the better solution than before are allowed to leave
behind a pheromone trail for others to follow. While an
ant traces a single path (or vector), an element is selected
probabilistically depending upon two factors.
• Pheromone concentration
• Desirability function (fitness function) which is
problem specific greedy heuristic to aid in search
for good solution
The fitness function in our case is FRBS, which
determines the throughput against any selected power
vector. This is shown in fig-5 while mathematically, it
can be written as;
1
2
1 1
1
1
1 1
(log ( )) ( ) (8)
1
( , )
N
k
k
N N
k k k
k k
N
k k k
k
R R
N
M r MCP
N N
FRBS p QoS
N
α
=
= =
=
=
= =
=
∑
∑ ∑
∑
In this equation the final p’s are determined by ACO,
with the help of that FRBS selects the optimum
modulation code pairs, that eventually results in an
enhanced bit rate.
B. Simulated Annealing
Simulated annealing (SA) is a global optimization algorithm
in which the concepts of statistical mechanics and
combinatorial optimization are combined. It was developed
by Kirkpatrick et al. in 1983[22]. 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.
SA also uses the same fitness function shown in fig-5
and described in Equ. (8). 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,
optimum vector is found that gives the highest throughput.
Figure 5. Fitness function Block
VII. RESULTS
In this section proposed scheme is compared with other
schemes. Table-1 contains the parameters used to perform
the simulations.
Table-1 Simulation Parameters
Sr. Parameter Value
1 Number of Subcarriers N=Number
of Ants
1024
2 Fitness Function for ACO and SA Fuzzy Rule Base System Fig-5
3 ACO and SA iterations 50
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, ACO-FRBS
8 Parameters being adapted Code rate, Modulation and power
In fig-6, performance of ACO-FRBS assisted adaptive
coding modulation and power (ACMP) scheme is
investigated for various target bit error rates that is 10e-2
(low) to 10e-5 (high). At 25dB throughput approaches
5.5bits/s/Hz, and as target BER is becoming stringent the
throughput is being compromised such that at very high
value of QoS it approaches to 3.5bits/s/Hz.
In fig-7 to fig-10, proposed scheme is compared with
SA-FRBS based ACMP [23] and FRBS assisted ACM [13]
scheme with fixed power case for different target BERs. In
fig-7, for SNR range of (0dB to 20dB) ACO-FRBS assisted
scheme significantly performs better than SA-FRBS and
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

232013 International Conference of Soft Computing and Pattern Recognition (SoCPaR)
fixed power schemes, in terms of throughput. While, for
SNR greater than 20dB ACO-FRBS and SA-FRBS perform
is nearly same but still it is far better than the fixed transmit
power case. In this case target BER was 10e-2.
In fig-8, results are almost similar that of fig-8 as far as
SNR range is till 20dB, however, for SNR>20dB SA-FRBS
is relatively better than ACO-FRBS scheme, and at 30dB
this difference approaches 0.5bits/s/Hz.
In fig-9, results are quite interesting, for SNR below or
equal 20dB, ACO-FRBS is far better than SA-FRBS and
fixed power scenario, but for SNR values between 23dB to
27dB SA-FRBS becomes up and for SNR values above
27dB both approaches have same performance.
In fig-10, however, ACO-FRBS assisted ACMP
outperforms compared to SA-FRBS assisted ACMP and
fixed power case with only adaptive coding and modulation
for all cases of SNR except in the vicinity of 25dB SNR both
schemes become close in terms of throughput.
Particularly, in this case, performance of SA-FRBS becomes
equal to the fixed power case at an SNR value of 15dB and
below 5dB but for rest of the values of SNR, SA-FRBS
outperforms compared to fixed power case.
Figure 6. Comparison of proposed scheme with QoS=10e-2 per subcarrier
Figure 7. Comparison of proposed scheme with QoS=10e-3 per subcarrier
Figure 8. Comparison of proposed schemes with QoS=10e-4 per subcarrier
Figure 9. Comparison of proposed schemes with QoS=10e-4 per subcarrier
Figure 10. Comparison of proposed scheme for different target BER
VIII. CONCLUSIONS
In this paper Ant Colony Optimization with a Fuzzy Rule
Based System (ACO-FRBS) is proposed for adaptive
coding, modulation and power (ACMP) in OFDM systems.

Also the proposed scheme is compared with Simulated
Annealing and fuzzy rule based system (SA-FRBS) assisted
ACMP scheme. From the simulation results it is deduced
that ACO-FRBS scheme performs significantly better than
SA-FRBS and fixed power schemes in terms of throughput
for SNR values below 20dB. For above 20dB SNR values,
both schemes perform somewhat identically, except some
minor fluctuation in some cases. However, it was apparent
from the simulations that adaptive power based schemes
outperform compared to fixed power scheme.
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 terms of rate
enhancement compared to its fixed power variant.
Performance is measured for different quality of service
demands (target BER) per subcarrier in OFDM system.
ACKNOWLEDGEMENTS
This research work was partially supported by Barani
Institute of Information Technology (BIIT), Rawalpindi,
Pakistan.
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] Daniels, R.C.; Caramanis, C.M.; Heath, R.W., "Adaptation in
Convolutionally Coded MIMO-OFDM Wireless Systems Through
Supervised Learning and SNR Ordering," Vehicular Technology,
IEEE Transactions on , vol.59, no.1, pp.114,126, Jan. 2010.
[19] Wahls, Sander; Poor, H.Vincent, "An outer loop link adaptation for
BICM-OFDM that learns," Signal Processing Advances in Wireless
Communications (SPAWC), 2013 IEEE 14th Workshop on , vol., no.,
pp.719,723, 16-19 June 2013.
[20] Alberto Rico-Alvarino, Robert W. Heath Jr.: Learning-Based
Adaptive Transmission for Limited Feedback Multiuser MIMO-
OFDM. CoRR abs/1306.0502.
[21] M. Dorigo, V. Maniezzo and A. Colorni,. "Ant System: Optimization
by a Colony of Cooperating Agents", IEEE Transactions on Systems,
Man, and Cybernetics–Part B, 26 (1), pp. 29–41, 1996.
[22] S. Kirkpatrick, C. D. Gelatt, and M. P. Vecchi, “Optimizationby
simulated annealing,” Science, vol. 220, no. 4598, pp. 671–680, 1983.
[23] Atta-ur-Rahman, Qureshi I.M, Salam M.H., Naseem M.T., “Efficient
Link Adaptation in OFDM Systems using a Hybrid Intelligent
Techniques,” 13th
International Conference on Hybrid Intelligent
Systems, Hammamet, Tunisia, Dec. 04-06, 2013.

paper2

More Related Content

What's hot (20)

Similar to paper2 (20)

paper2