International Journal of Electrical, Electronics and Computer Systems (IJEECS)

________________________________________________________________________________________________

Analysis of BER for transmission of image using QPSK and 16-QAM

1
S. N. Kulkarni, 2J. A. Desai
Department of E&TC Engineering, NBN Sinhgad School of Engineering, Ambegaon (Bk.)Pune, India
Email: [email protected], [email protected]

is used. Eb (bit energy) is signal power S times the bit

Abstract— Bit Error Rate is a critical issue in wiredand
wireless communication. This paper analyzes and assesses time Tb. N0 (noise power spectral density) is noise
Quadrature Phase Shift Keying (QPSK) and 16-QAM power (N) divide by bandwidth (W). Since bit time and
(Quadrature Amplitude Modulation) modulation bit rate Rb are reciprocal, we can replace Tb with 1/Rb.
techniques with respect to Bit Error Rate in presence of [2]
Gaussian, Rayleigh, and Rician noise. The primary aim of
any communication system is to receive data with
minimum errors as errors degrades system performance.
This paper focusses on the effect of different types of noise
for above mentioned modulation schemes under AWGN In Digital communication the number of bit errors is the
channel.The complete system is implemented in MATLAB number of received bits of a data stream altered due to
and MATLAB Simulink.The results obtained from the noise, interference, distortion or bit synchronization
Simulink models are used for the transmission of an image. errors. Bit error rate/ ratio (BER) is the number of bits in
Keywords— BER, AWGN, Gaussian, Rayleigh, error per number of transferred bits in total during a
Rician,MATLAB Simulink studied time interval. BER is a unit less measure,
expressed often as a percentage. The performance of
I. INTRODUCTION each modulation technique is measured by calculating
In digital modulation, digital symbols are converted into the BER in presence of different types of noise.
waveforms that are compatible with the characteristics III. DIGITAL MODULATION SCHEMES
of the channel. Bandpass modulation converts
information signal to sinusoidal waveform for digital A digital signal can modulate amplitude, frequency, or
modulation. Coherent and non-coherent phase of sinusoidal carrier wave. If the modulating
modulation/demodulation is the basic types of bandpass waveform consists of NRZ rectangular pulses, then the
modulation. Binary Phase Shift Keying (BPSK), modulated parameter will be switched or keyed from
Quadrature Phase Shift Keying (QPSK), and Quadrature one discrete value to another. This results into three
Amplitude Modulation (QAM) are specialized formats basic types of digital modulation schemes namely
of these techniques [3]. In an ideal channel the Amplitude Shift Keying (ASK), Frequency Shift Keying
transmitted signal from the transmitter will pass through (FSK), and Phase Shift Keying (PSK) [4].
channel upto the receiver, where it is demodulated to get This paper deals with the types of PSK viz. BPSK,
a perfect representation of the original signal. However DPSK, QPSK and QAM. This section will describe
in reality the received signal consists of mixture of these methods in detail.
attenuated and reflected version of the transmitted signal
[2]. In addition to these, the channel adds various types 1. BPSK: In BPSK, the carrier gets 0 or 180o
of noise to the signal. This affects the Bit Error rate of phase shift with respect to two different voltage levels of
the system. binary modulating signal. If the sinusoid is of amplitude
A, it has a power so that The transmitted signal is
II. BIT ERROR RATE given by
The quality of transmission is decided by parameter,
Signal to Noise ratio (SNR) in analog and by Bit Error
Rate

Where b(t) is a stream of binary digits with voltage

(BER) in digital. In digital communication, the ratio levels [6][7].
Eb/No, a normalized version of SNR as a figure of merit

________________________________________________________________________________________________
ISSN (Online): 2347-2820, Volume -5, Issue-1, 2017
96
 International Journal of Electrical, Electronics and Computer Systems (IJEECS)
________________________________________________________________________________________________

2. QPSK: When a data is transmitted using BPSK undeviated position of the state, the radial distribution of
technique the channel bandwidth required is 2fb. The the noise is used to add discrete time white Gaussian
QPSK technique reduces that bandwidth to fb. It is a noise in the channel
multilevel phase modulation. In this two successive bits
3. Rician Noise: Rician noise is signal-dependent
in a bit stream are combined together to form a message
and consequently separating signal from noise is a
and each message is represented by distinct value of
difficult task. It is problematic for low signal-to-noise
phase shift of a carrier. The QPSK signal is represented
ratio
as. Since there are 4 phases it is called as 4-PSK or
Quadrature PSK systems [6][7]. B. AWGN channel
Incommunication theory it is often assumed that the
transmitted signals are distorted by some noise. The
most common noise to assume is additive Gaussian
noise, i.e. the so called Additive White Gaussian Noise
channel, AWGN. AWGN is the channel in which noise
affects each transmitted symbol independently in the
3. QAM: QAM improves the noise immunity of detection process of it. Such a channel is called
the system by allowing the signal vectors to differ, not memory-less channel. The term additive indicates that
only in their phase but also in amplitudes. It utilizes the noise is simply superimposed or added to signal that
carrier phase shifting and synchronous detection to there are no multiplicative mechanisms at work. This
permit two DSB signals to occupy the same frequency channel is linear and time-invariant and its frequency
band. The two DSB signals are orthogonal to each other. response is flat for all the frequencies.
V.MATLAB SIMULINK MODELS
This section describes the simulation models of various
digital modulation techniques. A Bernoulli Binary
Generator feeds into digital modulation techniques
IV. NOISE AND CHANNELS IN (BPSK, QPSK, and QAM) used for transmission. To
COMMUNICATION SYSTEMS analyze the effect of noise, the modulated signal along
with the Rayleigh, Gaussian and Rician noise is
A. Noise in Communication Systems
transmitted on the AWGN channel. The received signal
The term noise refers to unwanted electrical signals that is demodulated using various demodulation techniques
are always present in electrical systems. The presence of and is used to calculate the bit error rate for transmission
noise superimposed on a signal tends to obscure the process. The BER is calculated by using the Monte
signal which limits the receiver’s ability to mask correct Carlo simulations in MATLAB Simulink Tool. The
symbol decisions, and thereby limits the rate of models shown below indicate the effect of Gaussian
information transmission [2]. Noise on the BER of BPSK, QPSK and 16 QAM. The
BER for the effect of Rician and Rayleigh noise is
Contaminating noise in signal transmission usually has
calculated by replacing the Gaussian Noise block in
an additive effect in the sense that noise often adds to
MATLAB Simulink by Rician noise and Rayleigh noise
the information-bearing signal at various points between
blocks respectively.
the source and the destination. For the purpose of
analysis, all the noise will be lumped into one source
added to the signal in the AWGN channel. So, in this
paper, effect of various noise on BER of different
modulation schemes have been studied. The various
sources of noise used for this system are mentioned
below:
1. Gaussian Noise: Various types of noise sources
are gaussian and have a flat spectral density over a wide
frequency range. Such a spectrum has all frequency Fig. 1: Effect of Gaussian noise on BPSK
components in equal proportion and is therefore called
white gaussian noise otherwise it is non-white gaussian
noise. The gaussian noise generator block
2. Rayleigh Noise: In digital communication, all
are interested in the two dimensional noise. The noise
can be characterized in two ways. A three dimensional
picture is given by the product of two orthogonal
gaussian distributions with the same standard deviation.
Alternately, with the polar coordinates centered on the Fig.2. Effect of Gaussian noise on QPSK
________________________________________________________________________________________________
ISSN (Online): 2347-2820, Volume -5, Issue-1, 2017
97
 International Journal of Electrical, Electronics and Computer Systems (IJEECS)
________________________________________________________________________________________________

Fig.3. Effect of Gaussian noise on QAM

The Simulink models shown above are built to analyze
Fig.4. Received image using QPSK and 16-QAM for
the effect of different types of noise (Rayleigh,
SNR=5
Gaussian, and Rician) on BER of BPSK, QPSK and
QAM. The BER obtained for those Simulink models
discussed below. Eb/N0 = 1and samples per
frame=1000, are tabulated as follows:
TABLE I. EFFECT OF NOISE ON BER
BER
BER BER with with BER with
Type of withou
t Rayleigh Gaussian Rician
Modulation Fig.5. Received image using QPSK and 16-QAM for
noise Noise Noise Noise SNR=10
BPSK 0.048 0.479 0.496 0.465

QPSK 0.245 0.723 0.745 0.726

QAM 0.538 0.852 0.867 0.847

VI. RESULTS AND DISCUSSIONS

For the MATLAB Simulink Models, discussed above Fig.6. Received image using QPSK and 16-QAM for
for Eb/N0 = 1 and samples per frame=1000, BER SNR=15
is tabulated asshown. From the given result TABLE I it
is observed that 16-QAM is least affected by Noise. It
can be observed that the difference between the BER of
QPSK and 16- QAM without noise and BER with Noise
is very less. So these modulation techniques are used for
the transmission of image over AWGN.
If QPSK and QAM are used for the transmission of
image the errors in the received image are less as BER is
directly proportional to the errors introduced in the
given data.The obtained results with Simulink model Fig.7. Received image using QPSK and 16-QAM for
helped to select modulation technique for transmission SNR=20
of image.The QPSK and 16-QAM modulation
techniques are used to transmit an image.The same It can be observed that less the value of SNR, content of
image is transmitted using QPSK and 16 QAM by noise is more. From the above images it can be seen that
changing the values of Signal to Noise Ratio (SNR). for low values of SNR, the image transmitted using
QPSK is affected more than the image transmitted using
16 QAM.
The reception of images is not affected much for higher
values of SNR.
Therefore, the 16 QAM technique can be used for the
effective transmission of multimedia over AWGN.

________________________________________________________________________________________________
ISSN (Online): 2347-2820, Volume -5, Issue-1, 2017
98
 International Journal of Electrical, Electronics and Computer Systems (IJEECS)
________________________________________________________________________________________________

VII. CONCLUSION [4] A. Bruce Carlson and Paul B. Crilly,

Communication Systems, 5th ed., McGraw-Hill,
In this paper effect of different types of noise on BER of pp 663-692.
various modulation schemes have been analyzed.
Simulation study shows that the systems are least [5] Bernard Sklar and Pabitra Kumar Ray, Digital
affected by Rician Noise as the Rician PDF (Probability Communications, 2nd ed., Pearson, pp 168-236.
Distribution Function) provides a better overall fit to the [6] Herbert Taub, Donald L. Schilling and
data than gaussian PDF. The table shows that BER value GoutamSaha, Principles of Communication
of QPSK technique is more in presence of additional Systems, 3rd ed., Tata MaGraw Hill, pp 322-418.
noise. It is proved by transmitting the same image using
QPSK and 16-QAM modulation techniques, in which [7] B. P. Lathi and Zhi Ding, Modern Digital and
the noise is least affected on the image transmitted using Analog Communication Systems, 4th ed., Oxford
16 QAM. University Press, pp 520-525.

REFERENCES [8] Md. Golam Sadeque, “Bit Error Rate (BER)

Comparison of AWGN Channels for Different
[1] Nagaraju, C, and S SParthaSarathy,”Analysis and Type’s Digital Modulation Using MATLAB
Estimation of Noise in Embedded Medical Simulink”, ASRJETS, pp 61-71.
Images”,International Journal of Image Graphics
and Signal Processing,2015 [9] V.Chauhan, M. Arora and R.S. Chauhan,
Comparative BER Performance of PSK based
[2] Mrs.J.A.Desai and Mrs.S.N.Kulkarni, “Effect of Modulation Techniques under Multipath Fading ,
Noise on BER of BPSK, QPSK, DPSK, and Advances in Applied Science Research , 2011,
QAM Modulation Techniques”, IJRSI, vol 3, 2(4): 521 -524.
issue 7, pp. 71-74, July 2015.
[10] Haykin, Simon S. Digital communications. New
[3] PratimaManhas and Dr M.K Soni, “Ber Analysis York: Wiley, 1988
Of BPSK, QPSK & QAM Based OFDM System
Using Simulink”, IJEEE, vol 7, issue 2, pp. 54-
60, July-Dec 2015.


________________________________________________________________________________________________
ISSN (Online): 2347-2820, Volume -5, Issue-1, 2017
99