SIDELOBE SUPPRESSION AND PAPR REDUCTION FOR COGNITIVE RADIO MIMO-OFDM SYSTEMS USING CONVEX OPTIMIZATION TECHNIQUE

International Journal of Information Sciences and Techniques (IJIST) Vol.4, No.3, May 2014
DOI : 10.5121/ijist.2014.4309 67
SIDELOBE SUPPRESSION AND PAPR
REDUCTION FOR COGNITIVE RADIO
MIMO-OFDM SYSTEMS USING CONVEX
OPTIMIZATION TECHNIQUE
Suban.A1
, Jeswill Prathima.I2
, Suganyasree G.C.3
,
Author 1 :
Assistant Professor, ECE Department, Velammal College of Engineering and
Technology
Author 2
, Author 3
: Final Year students, ECE Department, Velammal College of
Engineering and Technology
ABSTRACT
Orthogonal Frequency Division Multiplexing (OFDM) is deployed to overcome the interference. However,
OFDM has a relatively large OOB emissions. In spectrum sharing approaches such as dynamic spectrum
access networks, the OOB power levels of secondary transmissions should be kept below a certain level, in
order not to interfere with primary transmissions. . The difficulties such as sidelobes and PAPR caused by
OFDM is reduced by convex optimization and PTS technique respectively. In this technique each OFDM
subcarrier is multiplied with a real-valued weight that is determined in order not to interfere with adjacent
users. The problem with the SW technique is involving a very complex optimization. We propose a heuristic
approach called convex optimization. It can achieve considerable sidelobe suppression while requiring
significantly less computational resources than the optimal solution. Implementation results prove that it
can be introduced for real-time transmissions. Optimizing the subcarrier weights and SINR is complex, for
which we use the technique of convex optimization. For reducing the PAPR we use Partial Transmit
Sequence (PTS) technique.
Index terms : OFDM (Orthogonal Frequency Division Multiplexing), PAPR (Peak Average Power
Ratio), OOB (Out Of Band), IFFT (Inverse Fast Fourier Transform).
I.INTRODUCTION:
OFDM makes efficient use of the spectrum by allowing overlap of channels. By dividing the
channel into narrowband flat fading subchannels, OFDM is more resistant to frequency selective
fading than single carrier systems. It eliminates ISI and IFI through use of a cyclic prefix. But still
the use of OFDM introduces the sidelobes and the PAPR effect. Sidelobes is caused by the
multiple subcarriers of OFDM . In OFDM the single broad band frequency is divided into large
number of parallel narrow band of frequencies. OFDM inbuilt scheme produces orthogonal
carriers by using the Inverse Fast Fourier Transform (IFFT). In addition to that IFFT is also used
to raise the frequency used in the baseband to that of transmittable high frequency. Thus this
reduces the interference between the carriers of nearer frequencies. Moreover the cyclic prefix
addition makes us to reduce the most important problem of the digital communication that is the
Inter Symbol Interference (ISI). By using the frequency response of sub-carrier used for
transmission, the amount of information for each subband can be altered. On the other hand these
narrow bands have less frequency selective fading. But performing IFFT in the OFDM block the
peak amplitudes occur in the signal.
Though OFDM has a very high advantage over other techniques the major shortcoming involved

68
here is the high Peak to Average Power Ratio. The transmit signals in an
OFDM system can have high peak values in the time domain since many subcarrier components
are added via an IFFT operation. So there is a need for the reduction in this detrimental factor .In
the PTS technique, an input data block of N symbols is partitioned into disjoint sub-blocks. The
sub carriers in each sub-block are weighted by a phase factor for that sub-block. The phase
factors are selected such that the PAPR of the combined signal is minimized. The sidelobes
caused by OFDM is reduced by multiplying OFDM subcarrier with real valued weight that is
determined in order not to interfere with adjacent users. The problem with the SW technique is
involving a very complex optimization that has to be performed for each OFDM symbol. It can
achieved through convex optimization where considerable sidelobe suppression while requiring
significantly less computational resources than the optimal solution.
II. EXISTING METHODS:
A. Sidelobe Suppression by Multiple-Choice Sequences (MCS)
A set of sequences ( ))
= ( (p), (p). . . . . . . . (p)) ; p = 1,2.....P , is produced from the
sequence d. For each sequence d(p)
the average sidelobe power, is denoted with A(p)
; p = 1,2......P
, is calculated. To determine A(p)
, a certain frequency range spanning several OFDM sidelobes,
called optimization range, is considered using discrete frequency samples. Recalling that the
spectrum of an individual subcarrier equals a si-function si(x) = sin(x)=x, A(p)
is given by,
( )
= 1/ {
( )
( ( − )) }
p =1,2...............P
where xn, n = 1,2...... N, are the normalized subcarrier frequencies and yk; k = 1, 2.....K, are
normalized frequency samples within the optimization range. The index Q of the sequence with
maximum sidelobe suppression is given by
Q = arg min A(p)
; p = 1, 2...... P
Thus the sequence ̅ = ( )
is chosen for transmission and output from the MCS unit. To enable
successful data detection, the received sequence has to be de-mapped onto the original sequence
at the receiver. The MCS set is constructed such that the knowledge about the index Q of the
selected sequence is suffcient to perform this de-mapping. Thus, the index Q is coded in bits,
passed from the MCS unit to the signalling channel, and sent to the receiver.

69
Fig 1: Block diagram of the MCS sidelobe suppression unit
For example, assuming an OFDM system with N subcarriers modulated with M-ary phase-shift-
keying (M-PSK) or M-ary quadrature amplitude modulation (M-QAM) symbols, the overhead
needed for the signalling information is
[log ( )]/(log ( ). + [log ( )])
which is negligible for large N and/or M. [ ]denotes the smallest integer greater than or equal
to x. At the receiver, an estimate ( )
of the transmitted sequence ( )
is obtained which is
transformed into an estimate of the original sequence d using the signalling information. Note
that the signalling information is the index Q which indicates that the sequence d(Q)
out of the
MCS set has been chosen for transmission . In the following several computationally effective,
but yet effcient algorithms steps to generate MCS sets are proposed and analyzed. The proposed
methods do not degrade the bit-error rate performance at the receiver and require only a slightly
increased signalling overhead.
III. PROPOSED METHOD
A. To reduce sidelobes
To overcome the large computational complexity problem, we propose a heuristic approach to
perform optimization with power constraints. The basic idea of this approach is checking the
contribution of each weighted subcarrier in the OOB regions for and .The one that
results in lower OOB emissions is chosen for transmission. To understand the proposed
algorithm, let us assume that only the subcarriers at the borders of each OFDM symbol are to be
weighted while the remaining subcarriers are kept unweighted. Let and are the numbers of
the weighted subcarriers to the left and to the right of OFDM spectrum . We assume that and
are used to reduce the OOB emission the left and right of OFDM spectrum respectively. This
assumption is reasonable as the subcarrier closer to the edge of OFDM spectrum have impact in
the OOB region . First of all, the total OOB emission due to the unweighted subcarrier (N- -
) can be obtained from,
∑ ( )
The total OOB emissions will be only calculated at certain points (frequencies). Each subcarrier
has sidelobes in the OOB regions and the highest peak in the OOB region has the most significant
contribution in the OOB emissions. Therefore, the OOB emissions are calculated only at the
highest peak point of each subcarrier in the OOB region. It is assumed that each weighting factor
can be either or .The weighting factor are determined one after other by calculating
the OOB emission using and . Then the result in lower OOB emission is chosen for

70
transmission.
For illustration, consider weighting subcarriers. Define as the frequency domain
representation of the ith weighted subcarrier at the location of the highest peak in the OOB
region. Defined as the contribution of the ith weighted subcarrier at while is defined as
the total OOB emission at before assigning . The proposed algorithm can be described as
follows:
a) Calculate the OOB emissions due to the unweighted subcarrier at in OOB region.
b) Set i=1.
c) Calculate the OOB emission using and as:
= + ( ) ( )
= + ( ) ( )
d) If( < ) , then = . Otherwise, =
e) Calculate the total OOB emission at the location ,(i + 1 ≤ k ≤ ) by adding the
contribution of the ith weighted subcarrier using:
= + ( ) ( )
f) Increment i.
g) If ( i ≤ ). Go to step c.
The same procedure can be applied for the remaining subcarriers taking into account that these
subcarriers will be weighted to reduce the OOB emissions to the right of OFDM spectrum.
B. To reduce PAPR
This system is the incorporation of all the above mentioned techniques. The bandwidth is been
conserve d by the M-ary modulation techniques. The errors are minimized and optimized data
speed is achieved using MIMO antenna without any compromise for the bandwidth using OFDM.
In the below system, the signal is M-ary modulated and transmitted in MIMO antenna using
OFDM techniques where the factor of PAPR reduction is also considered for a better
performance than the present existing models
Fig 2 :The block diagram representing the combination of all the techniques is shown.
70
transmission.
follows:
b) Set i=1.
= + ( ) ( )
= + ( ) ( )
= + ( ) ( )
f) Increment i.
B. To reduce PAPR
70
transmission.
follows:
b) Set i=1.
= + ( ) ( )
= + ( ) ( )
= + ( ) ( )
f) Increment i.
B. To reduce PAPR

71
C. Combined model
The QAM modulated signal is converted from serial to parallel subject to the changes made by
subcarrier weighting. We select a minimal value for the weight vector based on the basic criterion
on SINR. This when done manually is a very tedious process, thus we do it using the technique of
convex optimization. OFDM has a main disadvantage of OOB emissions. To reduce interference
caused by OOB emissions, Cancellation Carriers are used. They are carriers added on either side
of the OFDM spectrum which can be calculated and cancelled on the receiver side. Then normal
process of OFDM is done by applying IFFT. The PAPR effect will affect the efficiency of the
communication system and thus we use Partial transmit sequence method to distribute the signal
in time domain. The parallel signals thus obtained are again converted into serial signal suitable
for transmission. Weighting is done accordingly during the transmission.
Fig 3: Proposed model
On the receiver side, the cyclic prefix is removed after converting it into a parallel signal. Then
the phase set is removed so that the signal is grouped in same time domain. The normal processes
of OFDM in a receiver like FFT are done and the cancellation carriers are calculated and
cancelled to obtain the original signal back. Then the signal obtained is converted again into a
serial signal and demodulated. The results have been analysed by comparing the original signal
with the received signal.
IV. RESULTS AND DISCUSSION
If the number of subcarriers increases, then the occurrence of the error decreases. This makes
OFDM more suitable for MIMO systems. The orthogonal carriers cause less interference in a
MIMO antenna that is closely placed. MIMO-OFDM gives more capacity than the conventional
MIMO in presence of multipath as shown.
71
C. Combined model
71
C. Combined model

72
Fig 4: BER vs. SNR for 4-QAM for various subcarrier with dimensions of MIMO as 2×2
In the above proposed system including the MIMO-OFDM schemes we find PAPR to be a factor that
need to be considered. So the extension of this system for a better performance will be possible by
reducing this PAPR to the minimal value possible by a suitable technique. Analysis can be done on
the system based CDF. The initial analysis here is done for an untrained system involving different
subcarriers without any technique to reduce PAPR. The estimation of PAPR for the system with
different subcarriers is shown below.
Fig 5: Effect of PAPR for Different Subcarriers using PAPR (dB) vs. CDF
The system with increased number of subcarriers shows an increase in the PAPR of the system. So a
power reduction technique is adopted to improve the performance of the proposed system. PTS
technique which is compatible with the above system is used.
PTS due to the moderate complexity and a better performance it is thus an attractive candidate for
PAPR reduction

73
Fig 6: Comparison of PAPR for system with PTS and without PTS techniques
An improved performance is thus obtained by using a Peak Power Reduction technique along
with the present system.
Fig 7 :Comparison of normal OFDM with Constellation Expansion method
This plot shows the spectrum of normal OFDM process compared with the constellation
expansion. It has reduced sidelobes. When subcarrier weighting is done, the sidelobe is even
more reduced.
V. CONCLUSION
The results shown above give an increase in the performance when an MIMO-OFDM system
uses QAM modulation along with PTS method implemented in it. By this way we can efficiently
communicate with low inter-channel interference and have minimal bit error rate. The high PAPR
is the detrimental aspect in this system. The SINR of the system is also optimized. Subcarrier
weighting and convex optimization technique is done to reduce sidelobe and interference. This
will form the Most Efficient Way of Communication.

74
REFERENCES
[1] Chirag Warty and Richard Wai Yu, “ Resource allocation using ASK, FSK and PSK modulation
techniques with varying M”, NAVSEA-Port Hueneme ,October 2010.
[2] Van Wyk, J. and Linde, L.,”Bit error probability for a M-ary QAM OFDM-based system,” IEEE
Transaction on Wireless Communication, 2007.
[3] Ye Li, Jack H. Winters and Nelson R.Sollenberger,”MIMO-OFDM for wireless communications:
Signal detection with enhanced channel estimation”, IEEE Transaction communications, vol 50, no.9,
September 2002.
[4] Min Shi, Claude D’Amours and Abbas Yongacoglu,” Design of spreading permutations for MIMO-
CDMA based on space-time block codes”, IEEE Communications letters, vol.14, no.1, January 2010.
[5] Pussadee Kiratipongvooth and Suvepon Sittichivapak,” Bit error probability of cooperative diversity
for M-ary QAM OFDM – based system with best relay selection”, IPCSIT vol.6, IA CSIT Press,
2011.
[6] Jiang and Hanzo,” Multiuser MIMO – OFDM for next-generation wireless systems”, Proceedings of
the IEEE Vol. 95, No. 7, July 2007.
[7] Simon Haykin,” Digital Communication”, Wiley India,2009.
[8] AndreasF.Molisch,”Wireless Communication”2ndedition, WileyIndia,2011.
AUTHORS
A. Suban, received B.E in the department of Electronics and Communication
Engineering from Anna university, Chennai and M.E in the discipline of Wireless
Technology from Thiagarajar College of Engineering, Madurai in 2011.
He is currently working as Assistant Professor in the Department of Electronics and
Communication Engineering, Velammal College of Engineering and Technology,
Madurai- 625009, Tamil Nadu, India. His area of interest in Signal processing mainly
focused on MIMO techniques with beamforming, OFDM and power control techniques
I.Jeswill Prathima pursuing Bachelor of engineering in Velammal College of
Engineering and Technology, Madurai-62 5009. Her area of interst is signal
processing
.
G.C Suganya Sree pursuing Bachelor o f engineering in Velammal College of
Engineering and Technology, Madurai-625009. Her area of interst is signal
processing.
74
REFERENCES
September 2002.
2011.
AUTHORS
processing
.
processing.
74
REFERENCES
September 2002.
2011.
AUTHORS
processing
.
processing.

SIDELOBE SUPPRESSION AND PAPR REDUCTION FOR COGNITIVE RADIO MIMO-OFDM SYSTEMS USING CONVEX OPTIMIZATION TECHNIQUE

More Related Content

Similar to SIDELOBE SUPPRESSION AND PAPR REDUCTION FOR COGNITIVE RADIO MIMO-OFDM SYSTEMS USING CONVEX OPTIMIZATION TECHNIQUE (20)

More from ijistjournal (20)

Recently uploaded (20)

SIDELOBE SUPPRESSION AND PAPR REDUCTION FOR COGNITIVE RADIO MIMO-OFDM SYSTEMS USING CONVEX OPTIMIZATION TECHNIQUE