BER and PAPR Reduction Using Companding Algorithms in OFDM

International Journal of Application or Innovation in Engineering& Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com

Volume 3, Issue 1, January 2014 ISSN 2319 - 4847

Volume 3, Issue 1, January 2014 Page 188

ABSTRACT
Orthogonal Frequency Division Multiplexing is an attractive technology for wireless communication. It offers high spectral
efficiency, power efficiency and multipath delay tolerance. The nonlinear characteristics of high power amplifiers are very
sensitive to signal amplitudes. OFDM is a multicarrier scheme in which number of carriers is added coherently, when they added
together it increases average power of signal amplitudes. High average power leads to distortion in operation of system. This
effect is known as peak to average power ratio (PAPR) is a major drawback of orthogonal frequency division multiplexing
(OFDM) systems. In literature many techniques were developed to reduce Bit error rate and PAPR. Among the various PAPR
reduction techniques, companding transform appears attractive for its simplicity and effectiveness. In this paper, evaluation of
performance of companding algorithms -law companding, Exponential companding, companding Airy function is compared
in terms of bit error rate and peak to average power ratio.
Keywords: Bit error rate, -law companding, Peak to average power ratio, nonlinear companding transform.
1. INTRODUCTION
Orthogonal Frequency Division Multiplexing is a multicarrier transmission technique, it divides available spectrum
into many carriers. OFDM uses spectrum much more efficiently by spacing the channels much closer to each other. This
system has better properties such as high spectral efficiency, robustness to channel fading, immunity to interference.
There are some obstacles in OFDM system. A major problem is that OFDM signal exhibits a very high peak to average
power ratio (PAPR). These large peaks cause saturation in power amplifiers, intermodulation amongst subcarriers,
increases out of band interference. Therefore it is necessary to reduce PAPR.
To reduce the peak to average power ratio several technique have been proposed such as clipping [3], [4], coding, peak
windowing, tone reservation [6]. But most of these technique are unable to achieve simultaneous a large reduction in peak
to average power ratio with low complexity. Among all these techniques the simplest solution is to clip transmitted signal
when its amplitude exceeds a desired threshold. Clipping highly non-linear process however, it produces significant out
of band interference (OBI). A good remedy for out of band interference is called companding. The technique soft
compress, rather than hard clips, the signal peaks causes far less out of band interference. The method was first
proposed in, which employed in classical - law transform and showed effective. Since then many different companding
transform with better performance have been published.
This paper organized as follows: section 2, presents companding algorithms such as law, Exponential, Companding
using airy function. In section three these algorithms compared with non-companded signal. We use bit error rate and
peak to average power ratio as a comparison parameter for companding algorithm. In section four we conclude.

2. Companding algorithm

Figure 1 Block diagram OFDM system with companding algorithm.
BER and PAPR Reduction Using Companding
Algorithms in OFDM

A. N. Jadhav
1
, M. V. Kutwal
2

1
Associate Professor, Electronics Department, D.Y. Patil College of Engineering & Technology, Kolhapur, India.

2
Assistant Professor, E&TC Department, D.Y. Patil Technical Campus, Talsande, Kolhapur, India



2.1 -Law Companding
Law is a simple but effective companding technique to reduce the peak-to-average power ratio of OFDM signal.
The idea comes from the use of companding in speech processing. Since OFDM signal is similar to speech signal in the
sense that large signals only occur very infrequently, the same companding technique might be used to improve OFDM
transmission performance [7],[8].
A QAM-OFDM system diagram is shown in Figure .1. The incoming bit stream is packed into x bits per symbol to
form a complex number where x is determined by the QAM signal constellation. For a real sequence output at the
IDFT, the complex input to the IDFT has Hermitian symmetry and is constrained as follows
(1)
Where k=0, 1, 2 (N/2)-1, and S(0)=0.

Suppose N is even and , the output of IDFT is

(2)
n=0, 1,2.,N-1
The law companding technique can be then introduced. The samples of OFDM signal s(n) are companded before it is
converted into analog waveform.
The signal after companding is given by

(3)
A is normalization constant, after D/A conversion the signal transmitted through channel. At the receiver end, received
signal first converted into digital form, the sampling result is

(4)
Where q is analog to digital conversion error and w is AWGN channel factor.
The expanded signal can be approximated as :

2.2 Exponential Companding
Exponential companding technique adjusts both large and small signals and can keep the average power at the same level
[9]. By transforming the original OFDM signal into uniformly distributed signals, the exponential companding schemes
can effectively reduce peak to average power ratio for different modulation subcarriers [9].Companding equation can be
given as: Let the power of amplitude of companded signal, have a uniform distribution in the interval [0,]. The
exponent is called degree of specific exponential companding scheme. The CDF of is simply
,0 (6)

Considering the phase of the input signals, the companding function h(x) is given by,

(7)

(8)
Where sgn(x) is the sign function. The positive constant determines the average power of output signals.

(9)


At the receiver side, the inverse function h(x) of is used in decompanding operation,

(10)
2 2. .3 3 Companding Using airy Function.
OBI is the spectral leakage into alien channels. Quantification of the OBI caused by companding requires the knowledge
of the power spectral density (PSD) of the companded signal. Unfortunately analytical expression of the PSD is in general
mathematically intractable, because of the nonlinear companding transform involved. Here we take an alternative
approach to estimate the OBI. Let (x) be a nonlinear companding function, and (t) =sin(t) be the input to the
compander. The companded signal (t) is: (t) =[(t)] = [sin(t)] . Since (t) is a periodic function with the same period
as (t), (t) can then be expanded into the following Fourier series:

(11)

Where the coefficients c (k) is calculated as:

(12)

Notice that the input x in this case is a pure sinusoidal signal, any (k) =0 for k >1 is the OBI produced by the
nonlinear companding process. Therefore, to minimize the OBI, (k) must approach to zero fast enough as k increases. It
has been shown that (k) k(m+1) tends to zero if y(t) and its derivative up to the m-th order are continuous, or in other
words, c(k) converges at the rate of k(m+1). Given an arbitrary number n, the n-th order derivative of y(t), Jny/Jtn, is
a function of Ji(x)/Jxi, (i =1, 2, , n), as well as sin(t) and cos(t), i.e.:
(13)

sin(t) and and cos(t) are continuous functions, Jny/Jtn is continuous if and only if Ji(x)/Jxi (i =1, 2, , n) are
continuous. Based on this observation : Companding introduces minimum amount of OBI if the companding function (x)
is infinitely differentiable. The functions that meet the above condition are the smooth functions. We now introduce a
smooth function, namely the airy special function. The companding function is as follows:

(14)

where airy() is the airy function of the first kind. o is the parameter that controls the degree of companding. [ is the
factor adjusting the average output power of the compander to the same level as the average input power:
(15)

Where E[] denotes the expectation. The decompanding function is the inverse of (x):

(16)

Where the superscript-1 represents the inverse operation. Notice that the input to the decompander is a quantized signal
with finite set of values. We can therefore numerically pre-compute 1(x) and use table look-up to perform the
decompanding in practice. Next we examine the BER performance of the algorithm. We can calculate BER by using
following equations. Let (t) denote the output signal of the compander, (t) the white Gaussian noise. The received signal
can be expressed as:

(17)

The decompanded signal (t) simply is:
(18)



Using the first order Taylor series expansion,

(19)

3. PERFORMANCE SIMULATION
From the given Equation shows that if (t) falls into the range of the decompanding function 1(u) where J1(u)/Ju
u=y(t) <1, the noise w(t) is suppressed, and if y(t) is out of the range, J1(u)/Ju u=y(t) >1 and the noise is
enhanced. Therefore, if the parameter o in is properly chosen such that more (t) is within the noise-suppression range of
1(u), it is possible to achieve better overall BER performance.

Figure 2 BER vs. SNR
Table 1: BER vs. SNR
Sr.No. SNR BER (non-
companded)
BER (-law companded) BER (Exponential
companded)
BER using airy
function
1 1 0.09 2.54 0.09 0.1
2 3 0.03 0.33 0.06 0.04
3 5 0.01 0.19 0.03 0.01
4 7 0.002 0.10 0.01 0.007
5 9 0.006 0.04 0.003 0.002

0 2 4 6 8 10 12 14
10
-1
10
0
PAPR
C
C
D
F

Exponential
Airy function
No companding
mu law

Figure 3 PAPR vs. CCDF
1 2 3 4 5 6 7 8 9 10 11
10
-4
10
-3
10
-2
10
-1
10
0
Performance analysis
-----EbNo
-
-
-
-
B
E
R

No companding
Airy function
Exponential
,mu law


Exponential companding possess lowest PAPR i.e. 2.8, airy companded has 4.1, and -law has 10.5. The non companded
signal has highest PAPR which causes problem in linear operation of OFDM system. It is observed that Non companded
signal has highest PAPR. -law & companding using airy function scheme gives improvement in PAPR, Exponential
scheme gives lowest PAPR as it works on high peak signals as well as small signals hence can keep average power at
minimum level. Also from table 1 airy companded signal has lowest BER than other companding techniques.
4. CONCLUSION
In this paper, performance of companding algorithms such as -law & Exponential, companding using airy function
algorithm is evaluated based on BER, PAPR. Airy companded signal has lowest BER than other companding techniques.
Non companded signal has highest peak to average power ratio. -law companded signal shows improvement in PAPR
than non-companded signal. Airy companded algorithm offers less PAPR than -law and non-companded signal.
Exponential companding scheme offers better improvement in PAPR among all these techniques as it adjusts both large
and small peaks. Thus by using airy companding scheme we can offer improved BER while reducing PAPR effectively.
References
[1.] A R. van Nee and R. Prasad, OFDM for Wireless Multimedia Communications. Boston, MA: Artech House, 2000
[2.] Ramjee Prasad ,OFDM for Wireless Communications Systems, Artech House, 2000
[3.] T. Jiang and Y. Wu, An overview: peak-to-average power ratio reduction techniques for OFDM signals, IEEE
Trans. Broadcast., vol. 54, no. 2, pp. 257-268, June 2008.
[4.] Luqing Wang and Chintha Tellambura A Simplified Clipping and Filtering Technique for PAR Reduction in
OFDM Systems, IEEE signal processing letters, vol. 12, no. 6, June 2005
[5.] Seung Hee Han, and Jae Hong Lee, Modified Selected Mapping Technique for PAPR Reduction of Coded OFDM
Signal, IEEE transactions on broadcasting, vol. 50, no. 3, September 2004
[6.] Abolfazl Ghassemi, and T. Aaron Gulliver, PAPR Reduction of OFDM Using PTS And Error-Correcting Code
Sub-blocking,IEEE transactions on wireless communications, vol. 9, no. 3, march 2010
[7.] X. Wang, T. T. Tjhung, and C. S. Ng, Reduction of peak-to-average power rat0 of OFDM system using a
companding technique, IEEE Trans. Broadcast., vol. 45, no. 3, pp. 303-307, Sept. 1999.
[8.] T. Jiang and G. Zhu, Nonlinear companding transform for reducing peak to- average power ratio of OFDM
signals, IEEE Trans. Broadcast., vol. 50, no. 3, pp. 342-346, Sept. 2004.
[9.] T. Jiang, Y. Yang, and Y. Song, Exponential companding technique for PAPR reduction in OFDM systems,
IEEE Trans. Broadcast., vol. 51, no. 2, pp. 244-248, June 2005
[10.] L.N. Bronshtein, K.A. Semendyayev, G. Musiol, and H. Muehlig, Handbook of Mathematics. Springer, 2007,
p. 422

AUTHOR
Prof. A. N. Jadhav received BE in Electronics from D.Y. Patil College of Engineering & Technology, Kolhapur in 1991,
ME degree in Electronics from walchand College of Engineering Sangli in 1997,(Ph.D. Scholar). He is currently
working as Associate Professor and HOD in D.Y. Patil College of Engineering & Technology, Kolhapur. He is a LM of
ISTE. His 33 international and 19 national research papers are published.

M.V. Kutwal completed BE from Terna College of Engineering Osmanabd in 2009 and doing M.E. ETC in D.Y. Patil
College of Engineering, Kolhapur and presently working as a Assistant Professor in D.Y. Patil Technical Campus,
Talsande, Kolhapur.

BER and PAPR Reduction Using Companding Algorithms in OFDM

Uploaded by

Copyright:

Available Formats

BER and PAPR Reduction Using Companding Algorithms in OFDM

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

BER and PAPR Reduction Using Companding Algorithms in OFDM

Uploaded by

Copyright:

Available Formats

International Journal of Application or Innovation in Engineering& Management (IJAIEM)

Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com

You might also like