Modified PTS Technique Of Its Transceiver For PAPR Reduction In OFDM System

Modified PTS Technique Of Its Transceiver For PAPR Reduction In OFDM System

1. Munmun Das

   Research Scholar MGM College of Engineering,

   Nanded(M.S),India.

2. Mr. Sayed Shoaib Anwar

Assistant Professor MGM College of Engineering,

Nanded(M.S),India.

1 N 1

j2 kn

Abstract

This work aims at developing a simulation model of

given by

j2 1 .

xn

Xk exp(

N

N

N

N

k 0

) where

Partial Transmit Sequence (PTS) Technique based on gray code of Orthogonal Frequency Division Multiplexing (OFDM) systems. As Traditional PTS

PAPR defined as maximum peak signal to the average peak signal.

P

technique need side information for transmitting, so it increases its complexity. This PTS technique based on gray code reduces its complexity to great extent as

PAPR 10 log10

2

2

1 T

peak [dB]

Pav

where

there is no need for side information and also worked well as a Peak to Average Power Ratio (PAPR)

Pav [x(t)] dt is the average power of the

T

T

0

suppression module compare to traditional PTS technique.

Keywords: PTS, PAPR, Gray code

1. Introduction

   transmitted signal and peak power.

   Ppeak

   max x(t) 2

   is the

   OFDM is a one of the popular multicarrier technique, came into existence from several decades. It has high data rate and high spectral efficiency. But it faces one major problem is high peak-to-average power ratio(PAPR) of transmitted signals, resulting in OFDM signals distortion in the nonlinear region of high power amplifier and high bit error rate [1]. To alleviate PAPR problem in an OFDM system, various techniques have been proposed such as selective mapping (SLM), partial transmit sequence (PTS) and active constellation extension (ACE).

2. PAPR Definition

   Let N denote the number of subcarriers used for

3. Proposed Partial Transmit Sequence Technique

   1. Conventional Partial Transmit Sequence Technique

      The PTS approach is well known method as a distortion less technique based on combining signal sub-blocks or clusters, which are multiplied by weighting factors. The PTS technique partitions the input block X of length N into V disjoint sub-blocks Xi of length N, i=1,2,···,V, which can be represented as v

      {X v =1, 2,,V}.

      V

      V

      Hence, X X v ,

      X

      X

      v

      v

      v1

      parallel information transmission and

      where X v [ X

      v , X v , X v

      N 1

      ] with

      Xk k

      1

      1

      Xk (0 k N 1) represent the k complex

      th

      or 0 (1 v V ) . Let b={bv e , v 1, 2,…V}be

      0

      0

      j v

      modulated symbol in a block of N information symbols. The outputs xn of the N-point inverse

      the set of phase factors which are applied to the subblocks X v . The substitute frequency domain

      discrete Fourier transform (IDFT) of

      Xk are

      signals are

      V

      v v

      v v

      X ' b X v , (b e jv , v 1, 2,…,V )…..(1)

      v1

      Note that these partial sequences are independently rotated by phase factors . Taking the IFFT of the above

      Let b1 and b2 be phase weighting sequences for

      equation and using the linearity property of the IFFT,

      generating the candidate signals

      y1 and

      y2 , and then

      the time domain partial transmit sequences can be expressed

      as x IFFT ( X ) b IFFT ( X ) b x

      as x IFFT ( X ) b IFFT ( X ) b x

      V V

      ' ' v v

      v v

      according to the rules 1 and 2, we can obtain the following formulae [1].

      v1

      v1

      Table 3.2.1: Phase weighting sequences

      No.	Bit Labeling	Gray Code
      b1	1000	1100
      b2	1001	1101
      b3	1010	1111
      b4	1011	1110
      b5	1100	1000
      b6	1101	1010
      b7	1110	1001
      b8	1111	1000

      No.	Bit Labeling	Gray Code
      b1	1000	1100
      b2	1001	1101
      b3	1010	1111
      b4	1011	1110
      b5	1100	1000
      b6	1101	1010
      b7	1110	1001
      b8	1111	1000

      (2) The objective is to optimally combine the V subblocks to obtain the time domain OFDM signals with the lowest PAPR. Without any loss of performance, one

      can set

      b1 1

      and observe that there are (V-1)

      subblocks to be optimized. Consequently, to achieve the optimal phase factor for each input data sequence (assume that there are W phase factors in the phase set),

      W v1

      combinations should be checked in order to

      obtain the minimum PAPR. Therefore, the search complexity for an optimum set of the phase factors increases exponentially with the number of subblocks [3].

   2. Modified Partial Transmit Sequence Technique

      In order to reduce the computational complexity of PTS, many papers have proposed effective solutions. PAPR Reduction of OFDM Signals Using a Reduced Complexity PTS Technique[2], Peak-to-Average Power Ratio Reduction of OFDM Signals Using PTS Scheme With Low Computational Complexity [3] used a low complexity phase weighting process is implemented,

      V

      V

      y1 b1,i xi x1 x2 x3 x4

      i1

      V

      V

      y2 b2,i xi x1 x2 x3 x4

      i1

      (3)

      ……….. (4)

      where the relationship between phase weighting sequences is considered and the computation for

      From above equations, it can be indicated that there is common term x x x . Let S = x x x ,

      candidate signals is simplified by making use of this

      inherent feature. These methods reduce the computational complexity to some extent, but the implementation of hardware is still so difficult. As these methods reduce the computational complexity to some extent, but the implementation of hardware is still so difficult. Aiming at this problem, the improved PTS approachs main idea is to reduce the correlation operation of the calculation by Gray code encoding the phase factors [1].

      Gray code is one of the popular code pattern

      1 2 3

      then y1 and y2 can be written as

      V

      V

      y1 b1,i xi S1 x4

      i1

      V

      V

      y2 b2,i xi S1 x4

      i1

      1 1 2 3

      (5)

      (6)

      for the structured light system. An n-bit Gray code is a

      From the above expressions we can find that we should

      kind of binary code whose adjacent code-strings differ

      calculate y1 first, and then the candidate signal

      y2 can

      only in one bit position. Take the nmber of sub-blocks V = 4 and the set of phase weighting factors W=2 is {1,

      -1} for example, all the phase weighting sequences are

      be easily obtained. On this basis, then be written as

      yk 1

      and

      yk can

      shown in Table 3.2.1.

      y b x b x b x

      S b x

      ..(7)

      V V

      V V

      k 1 k 1,i i k 1,i i k 1,m0 m0 i1 i1,im0

      k k 1,m0 m0

      V

      V

      k k , j i k , j i k ,m0

      k k , j i k , j i k ,m0

      y

      V

      V

      b x

      m0

      m0

      b x b x

      k k ,m0

      k k ,m0

      m0

      m0

      S b x

      Fig.3.4.2 PAPR performance comparison between the

      i1

      i1,im0

      ..(8)

      modified and the conventional PTS algorithm

      1. Reduced Computational Complexity PTS

         The improved PTS algorithm is mainly reflected in the calculation of reducing the amount of multiplication which reduces the hardware complexity. In the PTS algorithm, assuming that the number of sub- blocks is V, the number of phase factor is W, and the number of points in IFFT operation is N. Meanwhile, the computational complexity of traditional calculation PTS noted as O_PTS, the improved algorithm noted as R_PTS, we can obtain the following formulas [1].

         O _ PTS N (V 1) W v1

         Fig.3.4.1 shows that the BER performance of conventional and modified PTS are almost same but little bit differ from the theoretical value.

         As shown in the figure 3.4.2, when employ V=8, the PAPR performance increases 1.8dB compared with V=4, increases 2.7 dB when V=2. However, the computational complexity of V=8 is much larger than V=2 or V=4. Therefore, comparing the PAPR performance and the computational complexity, we divided entire data stream into 4, then the computational complexity of the final hardware

         R _ PTS N (V 1) N (WV 1 1)

         (9)

         implementation is lower and the PAPR performance can be achieved as well.

         Further simplify the ratio of the calculation:

         O _ PTS N *(V 1) N (W v1 1)

4. Design and Implementation

   R _ PTS N (V 1) *W v1

   (10)

   1. Modified PTS Simulink Model

      Eq.(10) shows that with the increase number of sub- blocks, the computational complexity reduces drastically. When employ sub-blocks defined value of V, the computation can be reduced to about 42%, comparing to the original PTS algorithm.

      1. Performance Analysis and Simulation Results

         Take computational complexity and PAPR performance into consideration, the simulation results is shown in figure 3.4.1. It can be seen through the MATLAB simulation, Gray code encoding PTS algorithm and the traditional PTS PAPR performance is almost in consistent, but the computation time of Gray code encoding PTS algorithm is greatly reduced. When We are using Conventional PTS approach to get a OFDM signal then it takes 12.004181 µs, while it only takes 10.516851µs Gray code encoding PTS approaches.

         Fig.3.4.1 BER performance of the theoretical,

         conventional modified PTS

         Fig. 4.1.1 A Simulink Transmitter Model

         This simulink model can be used in real-time application. The data is coded with any of the matlab coding and interfaced with the simulink model. So, reduction of the PAPR will be obtained in modified PTS model compare to OFDM model. This model is developed in order to give a comparison analysis of the performance with the OFDM model and modified PTS model. This system consisted of the Transmitter part consist of OFDM transmitter, PTS transmitter, Phase optimization, PAPR Calculation System and Receiver part consist of AWGN channel and BER Calculation system.

         Fig.4.1.2 Basic Receiver Simulink Model

         • Design Parameters

           Number of Symbols =15,000 Number of subcarriers = N=64 Modulation =M=16QAM Subblocks=V=4

           Number of phase factors=W=2=(1,-1) Symbol Length=T=1

           Symbol Energy=E=1

         • Phase Optimization

           This is the subsystem of phase optimization block. In this sub-system all combination of phases based on the binary representation of the phase are generated and the PAPR of all these combinations are computed. Then the smallest one will be selected for transmission. The simulink is usually used for the front-end system and is not as flexible as the m-file. Therefore, in this simulink only 8 combinations of different phases are considered as for the generation of the candidate signal formula is WV 1 .

         • PAPR Calculation

      It is used to calculate peak to average power ratio in dB.

      4.2. Results and Discussions

      Fig.4.2.1 shows the waveform of original ofdm signal and modified PTS

      Fig.4.2.2 shows the comparision of PAPR of original ofdm signal and modified PTS signal

      From the above figures, Fig.4.2.1 shows the waveform of original ofdm signal and modified PTS and .4.2.2 shows the comparision of PAPR of original ofdm signal and modified PTS signal, it can be concluded that PAPR of OFDM signal is greatly reduced.

      Fig.4.2.3 shows the square spectrum of input transmitt- ed signal and output received signal

      The PAPR of the original ofdm signal is 2.019dB and that of modified PTS signal is 1.688dB when simulation time is 15000.

      The signal to noise ratio is 10.31dB and bit error rate is 0.0034 of Modified PTS signal when simulation time is 15000.

5. Conclusion

   In Conventional PTS, the computation is high and need to transmit side information but when we use gray code

   than complexity is reduced and hardware can be implemented easily. Simulation results show that complexity is reduced to 42% of the original PTS when subblocks V=4.

   Matlab Stimulation show that Gray code Encoding PTS takes less time for encoding than Conventional PTS. We have consider AWGN channel, the bit error rate is almost same in both of them.

   By observing Waveforms, we conclude that high peak amplitude signals is greatly reduced in Modified PTS as compared to original OFDM waveforms. So there is no need of High Power RF Amplifier and also cost get reduced.

6. References:

[1]Liu Junjun Zhang Wei Yuan Zhu Ma Teng-Low complexity PTS algorithm based on gray code and its FPGA implementation, The Tenth International Conference on Electronic Measurement & Instruments ICEMI2011. [2]Seung Hee Han, PAPR Reduction of OFDM Signals Using a Reduced Complexity PTS Technique, IEEE Signal Processings Letters, Vol. 11, no. 11, November 2004

1. Peak-to-Average Power Ratio Reduction of OFDM Signals Using PTS Scheme With Low Computational Complexity by Jun Hou, Jianhua Ge, and Jing Li, IEEE Transactions on Broadcasting, VOL. 57, no. 1, March 2011

2. Wireless Communications with Matlab and Simulink:

IEEE802.1 Physial Layer

International Journal of Engineering Research & Technology (IJERT)

ISSN: 2278-0181

Vol. 2 Issue 2, February- 2013