1.[1 6] a novel papr reduction scheme based on selective mapping and a random-like coding with no explicit side information in ofdm

Journal of Information Engineering and Applications www.iiste.org
ISSN 2224-5758 (print) ISSN 2224-896X (online)
Vol 1, No.1, 2011

A novel PAPR reduction scheme based on selective mapping and
a random-like coding with no explicit side information in OFDM
Fateme Kargozar
Department of Electrical Engineering, University of Tabriz,Tabriz
PO box 5166-616471, Imam Khomeini Street, Tabriz, Iran
Tel: +989113438165 E-mail: Fateme.kargozar@gmail.com

Mehrbakhsh Nilashi
Department of Computer Engineering, Roudsar and Amlash Branch, Islamic Azad University,
Roudsar,Iran, PO box 1151-44815, Shohada Street, Roudsar, Iran
Tel: +98-911-344-3135 E-mail: Nilashidotnet@yahoo.com

Othman Ibrahim
Faculty of Computer Science and Information Systems, University Teknologi Malaysia,Johor
Alumni House, Universiti Teknologi Malaysia
Tel: +60127477698 E-mail: Othmanibrahim@utm.my

Mousa Barisami
Department of Computer Engineering, Roudsar and Amlash Branch, Islamic Azad University,
Roudsar,Iran
Tel: +98-911-243-8732 E-mail: barisamy.lahijan@gmail.com
Abstract
Orthogonal Frequency Division Multiplexing (OFDM) is a promising technique for high data rate and reliable
communication over fading channels. The main implementation drawback of this system is the possibility of
high Peak to Average Power Ratio (PAPR). In this paper, we develop a novel Selective Mapping (SLM) PAPR
reduction technique. In the novel proposed scheme, the alternative symbol sequences are generated by module 2
additions of data with the rows of cyclic Hadamard matrix with the same size, inserting the selected row’s
number to avoid transmitting any side information and specially using a random-like Irregular Repeat
Accumulate (IRA) encoder for both PAPR and Bit Error Rate (BER) better performance.
Keywords: IRA Codes, OFDM, PAPR, SLM method.

1. Introduction
Frequency Division Multiplexing (OFDM) has come to the forefront of technology over past decade because of
its robustness against multipath fading channels. It is an effective high-speed data transmission scheme without
using very expensive equalizers and for this reason, forms the basis of the physical layer of many broadband
high data rate technologies including Digital Subscriber Line (XDSL), WiFi (IEEE802.11a/g), WiMAX
(IEEE802.16a/e), and Digital Video Broadcasting (DVB).
One of the major drawbacks of OFDM systems is that the OFDM signal exhibits a high Peak to Average Power
Ratio (PAPR). Such a high PAPR necessitates the linear amplifier to have a large dynamic range which is
difficult to accommodate. On the other hand, an amplifier with nonlinear characteristics will cause undesired
distortion in band and out of band of the signals.
To deal with this problem, many methods have been proposed such as clipping and filtering, Active
Constellation Extension (ACE), Tone Injection (TI), Tone Reservation (TR), Partial Transmit Sequence (PTS),
Selective Mapping (SLM) and so some methods based on coding of the transmitting data.
Clipping and filtering can reduce PAPR of the OFDM signal but introduce the in-band clipping noise to it.

1|Page
www.iiste.org

Vol 1, No.1, 2011
Filtering is employed to remove the side-lobes generated by clipping, but it may also generate additional PAPR
[3, 4]. Active Constellation Extension (ACE), Tone Injection (TI) and Tone Reservation (TR) methods don’t
have the previous problems, but lead to use more power in the transmitter. Phase rotation is another approach to
reduce PAPR, including Selective Mapping (SLM) and Partial Transmit Sequence (PTS). These methods don’t
have the previous methods disadvantages such as BER performance degradation, because of out of band noise in
clipping-based methods, and high power consuming, in constellation extension-based methods, so these days
many researchers work on these methods optimization.
Our proposed scheme in this paper is based on a novel SLM method combined with a random-like coding of the
transmitted data that have no need to use side information for detecting transmitted data in the receiver.
To make this possible to understand our proposed scheme, we organized the rest of the paper as follows: The
problem of high PAPR of OFDM signal, structure of the conventional SLM scheme and the performance of IRA
encoder, as our random-like coding scheme is briefly defined in section 2. Section 3 introduces our proposed
technique and shows simulation results of it, in compare with the conventional SLM method. Finally, the
conclusions are drawn in section 4.

2. Bases of the proposed method
The OFDM signal sequence a = [a0 , a1 ,..., aN −1 ] using N = 2n subcarriers can be expressed as

N −1 j 2π
k
1 t
at = ∑ Ak e N
0 ≤ t < N −1
N (2.1)
k =0

Where A = [ A 0 , A1 ,..., A N −1 ] is an input symbol sequence, usually modulated by using Phase Shift Keying
(PSK) or Quadrature Amplitude Modulation (QAM) and t stands for a discrete time indexes.
The PAPR of the transmitted OFDM signal can be defined as

2
max at
0≤ t < N −1
PAPR (a ) = 2
E [a ] (2.2)

Where E denotes the expectation operator.
To generate U alternative symbol sequences, an input symbol sequences is multiplied by U different phase
sequences, each of length N:

Pu = [ Pu ,0 , Pu ,1 , ..., Pu ,N −1 ] 1 ≤ u <U (2.3)

The first phase sequence P1 is usually the all-1 sequence. Then, the alternative symbol sequences A u

A u = [AU , 0 , AU ,1 , ..., AU , N −1 ] 1 ≤ u <U (2.4)

are generated. We use the expression AU = A ⊗ PU to represent the component-wise multiplication in this form

A u ,n = A n Pu ,n o ≤ t < N −1 (2.5)

After U different alternative symbol sequences are transformed with IFFT, the OFDM signal sequence

2|Page
www.iiste.org

Vol 1, No.1, 2011
a u = IFFT(A u ) with the lowest PAPR is selected for transmission.
% %

If we assume that alternative symbol sequences are mutually independent, the Complementary Cumulative
Distribution Function (CCDF) expression for the SLM OFDM symbol can be given as

Pr(PAPR(a u ) > PAPR 0 ) = (1 − (1 − e − PAPR 0 ) N )U
% (2.6)
As it explained before, we can use random-like codes in the special form of them to reduce the PAPR of these
alternative symbol sequences.
The random-like codes offer capacity-achieving performance largely due to the random interleaver of the codes.
Here we utilize the inherent random interleaver in the random-like codes as a scrambler to obtain candidates of
independent data sequences. Shown in “Fig. 1”, a block of n − m information bits are first module2 adding with
the rows of cyclic Hadamard matrix, then inserted with m label bits and encode by a random-like IRA code.
These coded bits are modulated using 16-QAM modulation. IFFT is then applied to the modulated symbols. The
PAPR of the discrete OFDM signal is measured with four times oversampling and by enumerating the possible
sequences of inserted label bits before the encoding, we obtain different PAPR values. The selector selects the
one corresponding to the lowest PAPR to transmit.
Because of the random structure of the dense generator matrix in IRA codes, each information bit could affect
almost all the coded bits for a non-systematic code. Obviously, the non-systematic codes have better scrambling
effect. The systematic codes are also offer good randomization by employing the interleaver before modulation.
In the systematic IRA encoder, a block of information bits {d i } are encoded by an irregular repeat code with

di repeated ri times, where {ri : 2 ≤ ri < D } are the repetition degrees of {d i } and D is the maximum

repetition degree. The repeated bits are interleaved to obtain {u j } , and then encoded by an accumulator, given
by

a−1
x m+1 = x m + ∑ uam+i m = 0,1,..., M −1 (2.7)
i =0

Where x m represents parity nodes with initial states x 0 = 0 and a is the grouping factor. The length of the

L
n n = ri
parity bit is M = a , where i =0
. ∑

−L
The final coded bits {bi }iN=0 are the collection of the information bits
{x m }N =0
m .

Similar to LDPC codes, the IRA codes can be represented by a Tanner graph. Note that n is the total number of
edges connecting the information bit nodes and the check nodes. Define λi as the proportion of the edges
connected to the information bit node with degrees i and i = 2,..., D which satisfies

D

∑λ
i =2
i =1
(2.8)

And the rate of the codes is then given by

3|Page
www.iiste.org

Vol 1, No.1, 2011
n∑ i
λ
L i
R= = i
L + M n λi + λi n
∑ i ∑ a
i i

a ∑ λi
D

i
= i =0

1 + a∑
D
λi
i =0
i
(2.9)
3. Simulation Results of Our Proposed Scheme
Our proposed scheme contains cyclic Hadamard matrix to modify the transmitted data in U independent form.
Then, as it is shown in “Fig. 1”, we inserted these modified data with some lables that show which row of this
matrix adds with the original data so these lables have the ability to eliminate transmitting side information.
Then a random-like IRA encoder with a special generator matrix employed for both PAPR and BER better
performance in this novel scheme. 16QAM modulating and IFFT transformation are applied and the selection
part, selects the data with the lowest PAPR amount to transmit through the channel. Before transmitting, the
guard interval inserter applied on data to maintain the orthogonality of the transmitted data in the fading
channels in a cyclic extension method.
The structure of the receiver is shown in “Fig. 2”, while the PAPR and BER performance are demonstrated in
the latter figures in compare with the conventional SLM method. “Fig. 3”, “Fig. 4”, “Fig. 5” and “Fig. 6” shows
compares and results of Our Proposed Scheme.

4. Conclusion
This paper has introduced a novel PAPR reduction technique, in which, a random-like linear coding has been
used to reduce PAPR of OFDM system in a modified SLM method. The analytical derivation of the technique
has been given which describes the theoretical functionality of the technique. Simulation results show that using
a random-like coding such as IRA, because of getting the ability to select a special coding structure between
many random forms that gives lower PAPR has lead to better performance than the conventional SLM scheme.
In addition, the presented scheme has not only shown a significant reduction in PAPR, but also, implicitly an
improvement in BER without increasing the computational complexity and using any side information.
We use a polynomial to represent the repetition profile of an IRA code ensemble, i.e.
D
λ (x ) = ∑ λi x i −1
(4.1)
i =2

n
Denote r as the average repeat times of the bits, i.e. r = L .

Thus, the encoder of a systematic RA codes contains rL or aM additions. Normally r and a are much less
than the length of information bits but in general LDPC coding system we don’t have these conditions.
Therefore, the encoder of the systematic IRA code is much simpler than that of an LDPC code.

Acknowledgment
This work was supported by the Iranian Telecommunication Research Center (ITRC), Iran.

References
Saltzberg B.R., “Comparison of single-carrier and multitone digital modulation for ADSL applications,”. IEEE
Communications magazine, pp. 114-121, 1998.
REIMERS U., “Digital video broadcasting,” IEEE Communication magazine, pp. 104-110, 1998 .

4|Page
www.iiste.org

Vol 1, No.1, 2011
Ajit Jha Md. Sohel Mahmud an Sher, S. M. Shamsul Alam, Md. Tariq Hasan and Md. Mizanur Rahman,
“Reduction of Peak to Average Power Ratio (PAPR) in Orthogonal Frequency Division Multiplexing (OFDM):
a novel approach based on clipping and amplification,”. IEEE international Conference on Computer Sciences
and Convergence Information Technology, 2009
Tomohiro Noguchi, Khoirul Anwar a, Masato Saito and Minoru Okada, “Efficient PAPR for OFDM and
CI/OFDM system with iterative clipping,”. IEEE, 2008.
Byoung Moo Kang, Heung-Gyoon Ryu and Sang Burm Ryu, “A PAPR reduction method using new ACE
(Active Constellation Extension) with higher level constellation,” IEEE international conference on signal
prossesing and communication. 2007.
Tao Jiang and Yiyan Wu, “An overview: peak to average power ratio reduction techniques for OFDM Signals,”
IEEE tranzaction on broadcasting. 2008
Dae-Woon Lim, Hyung-suk Noh, Hyun-Bae Jeon, Jong-Seon No, and Dong-Joon Shin, “Multi-stage TR scheme
for PAPR reduction in OFDM signals. IEEE tranzactions on broadcasting. 2009
Kyn-Hong Kim, Hyun-Bae Jeon, Jong-Seon No and Dong-Joon Shin, “New SLM scheme for PAPR reduction
in OFDM Signals,”. International Symposium on Information Theory and its Applications (ISITA). 2008
Athinarayanan Vallavaraj, Brian G Stewart, David K Harrison and Francis G McIntosh. “Reducing the PAPR of
OFDM using a simplified scrambling SLM technique with no explicit side information,” IEEE international
conference on parallel and distributed systems. 2008.
Houshou Chen, Jyun-Jie Wang, Cheng-En Tu Hsing-Wang Chang, “PAPR reduction in OFDM systems based on
modified PTS algorithm with non-disjoint partition,” IEEE. 2009.
Guosen Yue and Xiaodong Wang, “A hybrid PAPR reduction scheme for coded OFDM,” IEEE tranzaction on
wireless communications, 2006.

5|Page
www.iiste.org

Vol 1, No.1, 2011

Figure 1. Transmitter of the proposed scheme

Figure 2. Receiver of the proposed scheme

0 -1
10 10
u=1
u=2
u=3 -2
10
u=4
pr (PAPR>PAPR0)

-1
10 u=8
BER

-3
10

-2
10 -4
10

-3 0 1 2 3 4 5 6 7 8 9 10
10
2 3 4 5 6 7 8 9 10 Eb/N0 (dB)
PAPR0 (dB)

Figure 3. PAPR performance of the Figure 5. BER performance of the conventional
conventional scheme scheme
0 -1
10 10
u=1
u=2
u=3 -2
10
u=4
pr (PAPR>PAPR0)

-1
10 u=8
-3
BER

10

-2
10 -4
10

-3
10 0 1 2 3 4 5 6 7 8 9 10
2 3 4 5 6 7 8 9 10
PAPR0 (dB)
Eb/N0 (dB)

Figure 4. PAPR performance of the proposed Figure 6. BER performance of the proposed
scheme scheme

6|Page
www.iiste.org

1.[1 6] a novel papr reduction scheme based on selective mapping and a random-like coding with no explicit side information in ofdm

More Related Content

What's hot (17)

Similar to 1.[1 6] a novel papr reduction scheme based on selective mapping and a random-like coding with no explicit side information in ofdm (20)

More from Alexander Decker (20)

Recently uploaded (20)

1.[1 6] a novel papr reduction scheme based on selective mapping and a random-like coding with no explicit side information in ofdm