201O International Conference on Networking and Information Technology

Performance Analysis of BPSK and QPSK Using Error Correcting Code

through AWGN

Suzi Seroja Sarnin Norasimah Kadri Aiza Mahyuni Mozi

Faculty of Electrical Engineering Faculty of Electrical Engineering Faculty of Electrical Engineering

University Technology MARA
40450 Shah Alam, Selangor
Darul Ehsan
suzis045@salam.uitm.edu.my
University Technology MARA
40450 Shah Alam, Selangor
Darul Ehsan
norasimah@salam.uitm.edu.my,
University Technology MARA
40450 Shah Alam, Selangor
Darul Ehsan
aiza mahyuni@salam.uitm.edu.my
_

Norfishah Ab Wahab, Nani Fadzlina Nairn

Faculty of Electrical Engineer Faculty of Electrical Engineer
University Technology MARA University Technology MARA
40450 Shah Alam, Selangor 40450 Shah Alam, Selangor
Darul Ehsan Darul Ehsan
fishah@salam.uitm.edu.my nanifadzlina@salam.uitm.edu.my

Abstract- This paper highlight the performance analysis of
BPSK and QPSK using error correcting code. To calculate the
bit error rate, different types of error correcting code were
used through an Additive White Gaussian Noise (AWGN)
channel. Bose- Chaudhuri-Hocquenghem (BCH), Cyclic code
and hamming code were used as the encoder/decoder
technique. Basically, the performance was determined in term
of bit rate error (BER) and signal energy to noise power
density ratio (EblNo). Both BPSK and QPSK were also being
compared in the symbol error capability known as t in which
expected that the performance is graded in response to the
increasing of value of t. All simulations were done using
MATLAB ru007b software. In general BCH
demonstrate better performance than Hamming code and
Cyclic code for both BPSK and QPSK.
Keywords: Additive White Gaussian Noise (A WGN) channel,
Bose-Chaudhuri-Hocquenghem code (BCH), Hamming code,
Cyclic code, energy per bit to noise power spectral density ratio
(EblNo), Bit Error Rate (BER), codeword length (N), message
length (K), error-correction capability (t).
I. INTRODUCTION
In digital communication, one of the most important
technical issues which are synchronization problem and the
error correction code used in this research are BCH code,
cyclic code and hamming code which is these code have
their unique advantage that suited to modifY this problem.
This is because the codes have the ability to recover the
synchronization problem since this code is self­
synchronizable [1]. Owing to this reason, the codes have
been choosing in order to study the performance of BPSK
and QPSK apply in communication system.
Since, this research is study the performance of phase
shift keying modulation, it is necessary to find the most
suitable channel that use to propagate the signal. Basically,
channel is fall into three types which are fading channels,
channels in which the noise stems from the others and
AWGN channel. As the author compared all the three type
of channels, the best suite channel for communication
system is AWGN channel. This is because in the practical
world is that AWGN never infinite in bandwidth. Thus, the
destruction process is successfully safe since the receiver or
measuring instrument has finite bandwidth [2].
In communications, the additive white Gaussian noise
(AWGN) channel model is one in which the only
codes
impairment with a linear addition of wideband or white
noise with a constant spectral density (expressed as watts
per hertz of bandwidth) and a Gaussian distribution of
amplitude. The model does not account for the phenomena
of fading, frequency selectivity, interference, nonlinearity or
dispersion [2]. However, it produces simple and tractable
mathematical models which are useful for gaining insight
into the underlying behavior of a system before these other
phenomena are considered [1].
This paper analyzes the performance of BCH code,
cyclic code and hamming code in AWGN fading channel
using BPSK and QPSK as the modulation scheme. The
study is mainly focuses on three type of error correction
code without considering the time taken in each of the
coded bits retransmission and other fading effects. This
paper also investigates more parameters which affect the
performance of the coding.
The Hamming Encoder block creates a Hamming code
with message length K and codeword length N. The number
N must have the form 2M-I, where M is an integer greater
than or equal to 3. Hamming codes can detect up to two
simultaneous bit errors, and correct single-bit errors; thus,
reliable communication is possible when the Hamming
distance between the transmitted and received bit patterns is
less than or equal to one. By contrast, the simple parity code
cannot correct errors, and can only detect an odd number of

978-1-4244-7578-0/$26.00 © 2010 IEEE 178

 2010 International Conference on Networking and Information Technology

errors [3]. Cyclic code are the subset of the class linear
codes of linear codes that satisfy the every codeword c=
(c1,... ,cn), then the word (cn,cl,... ,cn-l), obtained by a cycle
shifts of the elements of C, is also code word. The BCH
Encoder block creates a BCH code with message length K
and codeword length N. The input must contain exactly K
elements. The output is a vector of length N For a given
codeword length N, only specific message lengths K are
valid for a BCH code. No known analytic formula describes
the relationship among the codeword length, message
length, and error-correction capability [4].
This research purposely to analyze and simulate the
performance of BPSK and QPSK using three types of error
coding namely Hamming code, BCH code and Cyclic code.
Besides, it also compares both BPSK and QPSK in order to
yield the best performances in the term of BER as well as
EblNo. Furthermore, the performance of three types of error
coding in term of symbol error correcting capability (t) is
also highlighted for both BPSK and QPSK.
A comparison between simulated BER and theoretical
BER has been made as illustrated in Figure 3. The BER
value as EblNo varies from OdB to lOdB. At OdB, simulated
BER is 0.0771 and theoretical BER is 0.07856, which
comprised slightly difference values of 0.00155. At 10dB,
the simulated BER is 3.00E-06 and the theoretical BER is
3.87E-06, which comprised a difference of 8. 72E-07. Since
the difference between the simulated BER and theoretical
BER is insignificant, it can be observed that, the value of
the theoretical BER is comparable to the simulated BER.
B. Performance of a Communication System (using QPSK)
without Error Correcting Codes.
BER vs Eb/No without coding (QPSK)
l.OOE+OO
10 12

II. METHODOLOGY I.OOE-OI

�eER(simulated)
This research begins with the construction of the block ___ eER (t'leory)
diagram of the communication system. This block diagram 1.00E-02 1------------"'<----
play an important role in order to make sure this project is
satisfied with the communication system. The simulation
was divided into four parts: simulation without block codes,
1.00E-03
simulation with hamming code, simulation with cyclic code Eb/N.

and simulation with BCH code using the BPSK and QPSK Figure 2: The BER performance of QPSK without error
modulator/demodulator. During this research, all correcting codes.
simulations process representing the performance both
BPSK and QPSK is done using software known as Matlab By referring to Figure 2, at OdB, the simulated BER is
simulink. 0.3125 and theoretical BER 0.317310813 . The slight
difference between simulated BER and the theoretical BER
BER vs Eb/No without coding (BPSK) is 0.004811. While, at lOdB, the simulated BER is 0.001805
100E.00
and the theoretical BER is 0.001565402 , which comprised a
I difference of 0.00024. Still, the difference between the
0 10 12
simulated BER and theoretical BER is insignificant, thus, it
100E-01
can be observed that, the value of the theoretical BER is
comparable to the simulated BER.
100E-02

a:
C. Performance of communication system ofBPSK without
w 100E'()3 _BERlsimulated)
co and with error correcting codes using Hamming code
-+(BER theory)
100E'()4
BER Vs Eb/No between without codes
100E'()S and Hamming codes
1.00E .. OO

I,nOE-OI
1.00E-06
1.00E-02

1.00E-03
'"
-t--witoutcodes
1.00E-04
'"
�t-Iamming{7,4)
1.00E-OS
Figure 1: BER performance of BPSK without error correcting codes -a-Hamming(63.57)
1.00E-06

1.00E-07
III. RESULTS AND DISCUSSION
1.00E-08

A. Performance of a Communication System of BPSK

without Error Correcting Codes. Figure 3: BER vs (Eb/No) of a communication System
(using BPSK) without error correcting codes,

179
 2010 International Conference on Networking and Information Technology

Hamming code (7,4) and hamming code(63,57).
From the graph in Figure 3, the results demonstrate that
the BPSK using Hamming code (63, 57) has the lowest BER
compared to others. The BER for BPSK using Hamming
(63, 57) is 0.1042 at OdB and 2.00E-08 at lOdB. Thus, it can
be observed that BPSK using hamming (63, 57) demonstrate
better performance than others.
D. Peiformance of communication system of QPSK without
and with error correcting codes using Hamming code.
BER Vs Eb/No between without codes
and Hamming Codes
1.00E+OO
Figure 5: BER vs (EhINo) of a communication System (using BPSK)
without error correcting codes, BCH code (7,4, I) and BCH code(31,11,5).
By referring to graph in Figure 5, at 0 dB, the BCH
(31,11,5) coding demonstrated the highest value of 0.1481,
which is followed by BCH (7,4,1) and without coding.
While, at 10dB, BCH (31,11,5) exhibited the lowest BER of
O. The BER values of BCH (7,4,1) and without coding
exhibited values of 0.000005 and 3.00E-06 respectively.
Thus, it can be observed that BCH (31,11,5) has better
performance than BCH (7,4,1) and without code. The higher
error correction capability, t, will produce the better
performance in communication system.

J 10 12
F. Peiformance of communication system of QPSK
without and with error correcting codes using BCH
1.UUt:.-Ul
code
-+-without codes
�
1.00E-02
___ 1-lamming(7,4) BER vs E b/ N o without code and BCH
......-tlamming(G3,57) code for QPSK
1.00E..-00
1.00[-03
10 12

1.00E-01 E�!:!f:���_
1.00E-04
Eb/No
1.00E-02
�wlthoutcode

�BCH(7,4,l)
Figure 4: BER vs (EhINo) of a communication System 1.00[-03 -.-SCH (63.36.5)
(using QPSK) without error correcting codes,
Hamming code (7,4) and hamming code(63,57). 1.00E-04

By referring to graph in Figure 4, the performance of 1.00E-05 J

Hamming (7,4), Hamming (63, 57) and without codes are
observed in term of BER. At OdB, Hamming (7, 4) is the
highest BER compare to the others. At some point when
EblNo reach 8dB, the value of Hamming (63, 57) and
without code is almost the same which is 0.003848 and
0.004934 respectively. At lOdB, the value of BER of
Hammings (63, 57) demonstrated the least performance
compared to Hamming (7, 4) and without code. Thusm, it
can be observed that Hamming (63, 57) gives a better BER
values.
E. Peiformance of communication system ofBPSK without
and with error correcting codes using BCH code.
-'----------:
Eb '"""
:: ::-
N O-----
Figure 6: BER vs (EhINo) of a communication System (using QPSK)
without error correcting codes, BCH code (7, 4, I) and BCH code (63, 36,
and 5).
By referring to graph in Figure 6, at OdB, the highest
BER value is BCH (7,4,1) with 0.463 and the lowest BER is
without code which is 0.3125. At lOdB, the simulated BER
of BCH (63,36,5) demonstrated the lowest value compared
to the BER of BCH(7,4,1) and without code. The value BER
of BCH (63,36,5) at lOdB is 7.00E-05 and value of
BCH(7,4,I) and without code are 0.00189 and 0.001805
respectively. This proves that when the higher value of t, the
better performance of BER.

B E R vs Eb/No Between without code

G. Peiformance of communication system ofBPSK without
and BCH code
and with error correcting codes using cyclic code.
1.00E-01

1.00E-02

l.00E-03

�
�without code
1.00E-04
�
___ BCH (7,4,1)

l.00E-05 -a-BCH (31,11,5)

1.00E-06

l.00E-07

1.00E-(J8
Eb/No

180
 2010 International Conference on Networking and Information Technology

BER vs (Eb/No) of a communication System

BER vs Eb/No without code and Cyclic
(QPSK) for Hamming code, Cyclic code and BCH
code for BPSK code.
l.OOE+OO 1.OOE-tOO TI------�

10 11
l.OOE-01

l.OOE-02 1.OOE-Ol t----�:,.=_----

l.OOE-03 �without
ffi -+-qpsk BCH(7,4,1)
""
l.OOE-04 _cycliccode (7,4,1) 1.ooE02 _qpsk 8CH(63,36,5)

-a-cycliccode (15,7,2)
-a-qpsk cyclic code (31,H'::;)
l.OOE-OS
-qpslcrychc(/.4.1)
1.00E-03
l.OOE-06
-qpsk IlaTlming(7,4,l)

l.OOE-07 ...... qpsk Ha"mlng(63,57,1)

l00F-oL

Figure 7: BER vs (EbINo) of a communication System (using BPSK) 1.OOE-OS ..I.....--

Eb/No
without error correcting codes, cyclic code (7, 4, 1) and cyclic code (15, 7,
2).
Figure 9: BER vs (EbINo) of a communication System
(QPSK) for Hamming code, Cyclic code and BCH code_
By referring to graph in Figure 7, BPSK also has been
compared for various value of symbol error correcting
By referring to graph in Figure 9, the results show that
capability, t in term of BER. It can be observed that the
the best performance occurs when the communication
values of BER of EblNo from OdB to 6dB is almost the
system uses a BCH code with N=63, K=36 and t=5 with
similar to each other. Similarly, with value of t is being
QPSK modulator/demodulator.
compared for t equal to 1 and t equal to 2 respectively. The
fact that the higher value of t will cause the better
performance in term of BER that has been achieved for BER vs (Eb/No) of a communication System
Cyclic code (15,7,2) with 3.00E-07 at lOdB while 0.1091 at (BPSK) for Hamming code, Cyclic code and BCH
OdB.
code.
l.ooE+OO ,---��-�-�-��
H. Peiformance of communication system of QPSK without
l.U<Jt-<n
and with error correcting codes using cyclic code __ bpsk BCr(31,ll.l)
l.Wt.()1

l.Wt.-<h ....... bpsk 0Icl;c(7,4,1)

BER vs Eb/No without and Cyclic Code for QPSK

�
'" l.Wt.()L -bp,kqci;c(15,7,l1
m

l.00F-O\ �bp�k HdlUlllillg(7,4.1)

1.(lOF-Oh _bp�k Hdllunil1g(63,57,1)

12
1.00F-07 ...... bpsk BCr(7.4,1)

l.ODE-01 t-----"....,rIiIb.:,---
... ----
- l.OtlF-OR

�wi!h:>ut
� l.00E-02 +----------'�
_cyciiccodep,l,l)
Figure 10: BER vs (EbINo) of a communication System
(BPSK) for Hamming code, Cyclic code and BCH code_
-a-cycli::codc (31.21,5)

l.ODE-03 +----------\--- As in Figure 10, the results show that the best
performance occurs when the communication system uses a
BCH code with N=31, K=l1 and t=5 with BPSK
l.00[-04 -'-----------,-,----
- ---­
[b/No modulator/demodulator. In general, the BCH codes are
better than Hamming code and Cyclic code. This is because
Hamming codes and Cyclic code are capable of detecting
Figure 8: BER vs (EbINo) of a communication System (using QPSK)
without error correcting codes, cyclic code (7, 4, 1) and cyclic code (31, 21, and correcting single errors only whereas BCH codes are
5). capable of detecting and correcting multiple errors.

By referring to graph in Figure 8, it can be observed IV. CONCLUSION

that the smallest value of BER is when using Cyclic code
(31, 2 1, 5 ). The simulation shows that the performance of QPSK
and BPSK is dependent on several factors. The symbol
1. Peiformance Comparison between Communication error correcting, t, codeword length, N and message length,
System with BCH Code, cyclic code and with Hamming K. Based on the result obtained, it can be concluded that the
Code Using BPSK modulator/demodulator and QPSK best performance is graded when the value of t is increased.
modulator/demodulator. The best performance occur when communication system
use a BCH code with N=31, K=l1 and t=5 with BPSK

18 1
 2010 International Conference on Networking and Information Technology

modulator or demodulator. The higher of value of N, K and
t is better the performance and in general BCH codes are
better than Hamming code and cyclic code Hence, the
objective of this research is successfully achieved in which
is success to analyze and simulates the performance of
BPSK and QPSK using different types of error correcting
codes through A WO N channel. For instance, the order
phase shift keying modulation is determined the quality of
the modulation process and also determine the capacity data
that can be transfer. As the order of PSK is increase, the
performance will be degraded since the order of
constellation is more susceptible to noise. Besides, the
symbol-error correcting capability (t) also give some impact
to the performance of PSK since this research apply error
correction code as the encode/decode technique.
[15] M.A. Kousa and A.H. Mugaibel, "Puncturing effects on turbo
codes," IEE Pme. Commun. Vol. 149, No 3, June 2002
[16] M. Liinaharja, "Studies On the Performance of Some ARQ
Schemes," Helsinki University of Technology
Communications Laboratory Tech. Rep. T54, 2006
[17] S. Falahati, P. Frenger, P. Orten, T. Ottosson and A.
Svensson, "Convolutional Coding and ARQ Schemes for
Wireless Communications," in Proc. IEEE Nordic Radio
Symposium, Saltsjobaden, Sweden, October 1998, pp203-21O.
[18] P. Robertson, E. Villebrum and P. Hoher, "A comparison of
optimal and suboptimal MAP decoding algorithms operating
in the Log-domain", Int. Conf. on Communications, pp 1009-
1013, 1995
[19] S. Lin and D. 1. Costello, "Error Control Coding:
Fundamentals and Applications," 2nd ed. Prentice Hall, 2004.
[20] M. Kouraichi, O. B. Belghith, A. Kachouri and L. Kamoun,
"Evaluation of SOYA algorithm in Turbo code", IEEE
Transactions on Information Theory, pp. 659-663, March
2004.

REFERENCES

[I] http://en.wikipedia.org/wikilPhaseshift_keying#Quadrature�ha
seshift_keying_.28QPSK.29
[2] Vishakan Ponnampalam, Branka Vucetic, "Maximum
Likelihood Decoding Of Reed Solomon Codes," ISIT 1998,
Cambridge MA USA, August 16-August 21.
,

[3] MacWilliams, F. J. and Sloane, N.1. A. The Theory of Error­

Correcting Codes. Amsterdam, Netherlands: North-Holland,
1977. [E-book] Available at www.Wileyeurope.com.
[4] Jorge Castineira Moreira and Patrick Guy Farrell, Essentials of
Error Control Coding .Argentina: wiley and Son, 2006.[E­
book] Available at www.Wiley.com
[5] Andre Neubauer, Jurgen Freudenberger, and volken Kuhn.
Coding Theory Algorithms, Architectures and Application
.England: Wiley and son, 2007. [E-book] Available at
www.Wileyeurope.com.
[6] U. Elisabeth, "Hybrid ARQ Using Serially Concatenated Block
Codes for Real-Time Communication - An Iterative Decoding
Approach," Licentiate Thesis, Chalmers University of
Technology, Sweden, 2001
[7] K. Oteng-Amoako, 1. Yuan, S. Nooshabadi, "Selective Hybrid­
ARQ turbo schemes with various Combining methods in Fading
Channels," in Proceeding of Workshop on Modelling and
Optimisation on Mobile Ad Hoc Networking and Computing
(Wiopt03), France, 2003
[8] F. Chiti, R. Fantacci, F. Versaci, "Turbo soft combining hybrid
ARQ techniques: theory and application to 3G wireless
networks," in Annales des Telecommunications, 2005, v:60,
n:I-2, pp: 169-183
[9] L. Bahl, 1. Jelinek, 1. Raviv, and F. Raviv, "Optimal Decoding
of Linear Codes for Minimizing Symbol Error Rate," IEEE
Trans. on Information Theory, vol. IT-20, pp. 284-287, Feb
1974.
[10] G. D. Forney, "The Viterbi Algorithm," Proc. Of the IEEE, vol.
61, pp. 268-278, Mar 1973.
[II] Tsg-Ran WgI#17, "Performance Comparison of Hybrid-ARQ
Schemes," Motorola, Stockholm, Sweden, Tech. Rep.
TSGRI#17(00)1396-Adhoc#24-HSDPA, 2000.
[12] A. Banerjee, D. 1. Costello Jr., T. E. Fuja, "Performance of
hybrid ARQ schemes using turbo trellis coded modulation for
wireless channels," University of Notre Dame, 2002.
[13] T. M. Kim, H. Shin, and J. H. Lee, "A novel error detection
scheme for Turbo coded hybrid ARQ," in Proc. of the IEEE
VTC 2002-Fall, Vancouver, Canada, Sept. 24-28, pp. 2168-
2171, 2002.
[14] T. Rodrigues, R. D. Souza, M. E. Pellenz, "Hybrid ARQ
Scheme Based on Recursive Convolutional Codes and Turbo
Decoding", IEEE Transactions on Communications, 2008.

182