JOURNAL OF TELECOMMUNICATIONS, VOLUME 5, ISSUE 2, NOVEMBER 2010

24

Performance Analysis of an Efficient Wireless

Communication System in AWGN and Slow
Fading Channel
M. S Miah, Dr. M. M Rahman, T. K Godder and B. C Singh

Abstract— In this paper, first build up a wireless communication simulator including Gray coding, modulation, different channel
models (AWGN, flat fading and frequency selective fading channels), channel estimation, adaptive equalization, and
demodulation. Next, test the effect of different channel models to the data and image in receiver with constellation and BER (bit
error rate) plots under QPSK modulation. For Image data source, we also compare the received image quality to original image
in different channels. At last, give detail results and analyses of the performance improvement with channel estimation and
adaptive equalization in slow Rayleigh fading channel. For frequency selective fading channel, use linear equalization with both
LMS (least mean squares) and RLS (Recursive Least Squares) algorithms to compare the different improvements. We will see
that in AWGN channel, the image is degraded by random noise; in flat fading channel, the image is degraded by random noise
and block noise; in frequency selective fading channel, the image is degraded by random noise, block noise, and ISI.

Key Word—Slow fading, flat fading, frequency selective fading, channel estimation, LMS, RLS.

——————————
——————————

1 INTRODUCTION

M OBILE communications and wireless networks

have experienced massive growth and commercial
success in the recent years. However, the radio channels
pilot-based channel estimation with training data. The
equalization removes the effect of the wireless channel
and allows subsequent symbol demodulation. An adap-
in mobile radio systems are usually not amiable as the tive equalizer is a time-varying filter which must con-
wired one. Unlike wired channels that are stationary and stantly be retuned. A number of different algorithms can
predictable, wireless channels are extremely random and be employed for these modules. In our project, we use
time-variant. It is well known that the wireless multi-path LMS (least mean squares) and RLS (Recursive Least
channel causes an arbitrary time dispersion, attenuation, Squares). Digital communication systems operating on
and phase shift, known as fading, in the received signal. time varying dispersive channels often employ a signal-
Fading is caused by interference between two or more ing format in which customer data are organized in
versions of the transmitted signal which arrive at the re- blocks preceded by a known training sequence. The train-
ceiver at slightly different times. There are many diversity ing sequence at the beginning of each block is used to
techniques to address fading issue, such as OFDM, MI- estimate channel or train an adaptive equalizer. Depend-
MO, RAKE receiver and etc. However, it may be still ne- ing on the rate at which the channel changes with time,
cessary to remove the amplitude and phase shift caused there may or may not be a need to further track the chan-
by the channel if we want to apply linear modulation nel variations during the customer data sequence. Fig.1
schemes, such as the ones used in WiMAX. The function shows the flow chart of our MATLAB simulation which is
of channel estimation is to form an estimate of the ampli- used in this project.
tude and phase shift caused by the wireless channel from
the available pilot information. Channel estimation me-
thods may be divided into two classes: pilot based estima-
2 WIRELESS MOBILE COMMUNICATION SYSTEMS
tion and blind estimation. In our project, we will focus on In any communication system, there must be an in-
————————————————
formation source (transmitter), a destination (receiver)
and a medium to transmit information between the
• M. S Miah, Lecturer, Department of Information & Communication
transmitter and the receive. The block diagram of a basic
Engineering,Islamic University, Kushtia-7003, Bangladesh. communication system is given in Fig-1. Message source
• Dr. M. M Rahman, Professor, Department of Information & Communica- originates message such as human voice, a television pic-
tion Engineering,Islamic University, Kushtia-7003, Bangladesh. ture a teletype message or data. The message can be elec-
• T. K Godder, Associate Professor, Department of Information & Commu-
nication Engineering,Islamic University, Kushtia-7003, Bangladesh. trical and non-electrical. If it is not electrical, the source
• B. C Singh, Lecturer, Department of Information & Communication transducer will convert it into electrical signal. The
Engineering,Islamic University, Kushtia-7003, Bangladesh. transmitter may be consists of analog to digital converter,
data compressor, source encoder, channel encoder a
modulator or any other complicated subsystems.

© 2010 JOT
http://sites.google.com/site/journaloftelecommunications/
 25

ted power, infrastructure cost and complexity are low, TV

remote control, garage door opener and cordless phone
are few examples of non-cellular wireless communication
serving a small coverage area, for wide area coverage,
high power transmitter placed at a high elevation is used,
a single high power transmitter would be used for a large
city. The non-cellular mobile communication system is
also being used by small taxi-companies, utility compa-
nies, fire, police, medical and emergency personnel and
national operators of large vehicle fleets based on land
and sea. Wireless local area networks (WLANs) and the
evolving Bluetooth technology are also examples of non-
Fig. 1. Block diagram of basic communication system cellular wireless communication.

The receiver may be consists of demodulator, channel 2.2 Cellular wireless mobile communication
and source decoders data expender, digital to analog In the early mobile radio systems, the objective was to
converter or others. Receiver transducer converts the elec- achieve a large coverage area by using a single, high po-
trical signal to its original form- the message. Message wered transmitter with an antenna mounted on a tall
destination is the actual unit to which the message it sent. tower, as frequency reuse was not possible, the problem
The channel is the information transmission medium. of this system was that the user capacity was very low.
This medium can be of different types such as wire, a wa- For example, the Bell Mobile system in New York city in
veguide, an optical fiber or a wireless link. As the channel j1970s could only support a maximum if twelve simulta-
act as a filter, during the transmission of the signal (mes- neous calls over a thousand square miles. Thus the need
sage) through the channel, the signal can be distorted due of higher capacity with limited radio channel brought
to the attenuation and phase shift suffered by different into the cellular concept. Modern cellular concept began
frequency component of the signal. Noise will also be to appear in Bell system proposals during the late 1940s.
added with the transmitted signal during the transmis- in cellular case, a coverage area is divided into a large
sion of the signal through the channel. This noise is of number of small cells, each equipped with a low-power
random type and unpredictable. Interference from other transmitter and each transmitter provides services for a
users; faulty electrical equipments; automobile ignition small portion of the total geographic service area. The
radiation; fluorescent light, lightning, solar and interga- same frequencies could be reused in different cells. The
lactic radiation; thermal motion of electrons in conduc- most important properties of cellular architecture are as
tors; random emission, diffusion and recombination of below.
charged carriers in electronics devices are some sources of
channel noise. In case of wireless communication system,
 Low power transmitters per cell
the channel will be a radio link, which means free space  Small coverage area per cell
propagation is used in this case. There will be no physical
connection between the source and the destination. In  Frequency reuse
case of wireless communication two cases can arises as  Cell splitting to increase capacity
follows.
1. The source and the destination both are static, 2.3 Cellular Telephone System
i.e., they are fixed in position and not movable. A cellular telephone system provides a wireless con-
2. The source and the destination are not static, i.e., nection to the PSTN for any user location within the radio
either source and destination or both are mova- range of the system. Cellular systems accommodate a
ble. large number of users over a large geographic area, with-
in a limited frequency spectrum. Cellular radio systems
The second case where the source and the destination
provide high quality service that is often comparable to
can be moveable and radio link is used for communica-
tion, is termed as wireless mobile communication. that of the landline telephone systems. High capacity is
The wireless mobile communication can be of two types: achieved by limiting the coverage of each base station
transmitter to a small geographic area called a cell so that
the same radio channels may be reused by another base
1. Non-Cellular (i.e., signal cell) Mobile Communi- station located some distance away. A sophisticated
cation. switching technique called a handoff enables a call to pro-
2. Cellular Mobile Communication. ceed uninterrupted when the user moves from one cell to
another.
2.1 Non-cellular wireless mobile communication A basic cellular system which consists of mobile sta-
In non-cellular (single cell) mobile communication sys- tions, base stations and mobile switching centers (MSC),
tem, the coverage area is not divided into small cells as in the mobile switching center is sometimes called a mobile
cellular system. The coverage area can vary from a small telephone switching office (MTSO), since it is responsible
area to a wide one. For small coverage area, the transmit- for connecting all mobiles to the PSTN in a cellular sys-
 26

tem. Each mobile communicates via radio with one of the panies which can rapidly scan all of the possible FCCs to
base stations and may be handed-off to any number of determine the strongest channel at any time. Once finding
base stations throughout the duration of a call. The mo- the strongest signal, the cellular phone receiver stays
bile station contains a transceiver, an antenna, and control “camped” to the particular FCC. By broadcasting the
circuitry, and may be mounted in a vehicle or used as a same setup data on all FCCs at the same time, the MSC is
portable hand-held unit. The base stations consist of sev- able to signal all subscribers within the cellular system
eral transmitters and receivers which simultaneously and can be certain that any mobile will be signaled when
handle full duplex communications and generally have it receives a call via the PSTN.
towers which support several transmitting and receiving
antennas. The base station serves as a bridge between all
mobile users in the cell and connects the simultaneous 2.4 Problems in Cellular Mobile Communications
mobile calls via telephone lines or microwave links to the In a cellular system interference is the major limiting
MSC. The MSC coordinates the activities of all of the base factor in increasing capacity. Some of the interference
stations and connects the entire cellular system to the sources are for example. Other base station transmitting
PSTN. in the same frequency band, another mobile user in the
A typical MSC handles 100,000 cellular subscribers and same cell, a call in progress in a neighboring cell, impair-
5,000 simultaneous conversations at a time, and accom- ments caused by the propagation of radio waves, Etc. as a
modates all billing and system maintenance functions, as result. Different types of system interference are yielded
well. In large cities, several MSCs are used by a single in the network. Among these interferences the most im-
carrier. Communication between the base station and the portant are the following.
mobiles is defined by a standard common air interface
(CAI) that specifies four different channels. The channels Co-channel interference (CCI this type of interference
used for voice transmission from the base station to mo- is caused by the interference between co-channel cells
biles are called forward voice channels (FVC), and the (cells with the same frequency channel) due to the fre-
channels used for voice transmission from mobiles to the quency refuse. To reduce CCI, co-channel cells must be
base station are called reverse voice channels (RVC).The separated by a minimum distance to provide sufficient
two channels responsible for initiating mobile calls are the isolation due to propagation distance.
forward control channels (FCC) and reverse control
channels (RCC). Adjacent channel interference (ACI) This other type
of interference results when two frequency channels are
adjacent in the frequency spectrum and one of them is
leaking into the pass band causing interfering into the
adjacent channel.

Fig. 1. Magnetization as2a A

Fig. function
CelluarofSystem
applied field. Note that “Fig.” is

Fig. 3 Adjacent Channel

Control channels are often called setup channels be-
cause they are only involved in setting up a call and mov-
ing it to an unused voice channel. Control channels ACI is mainly aggravated by imperfect receiver filters.
transmitted receive data messages that carry call initiation This problem can be minimized with a careful filtering
and service requests, and are monitored by mobiles when and channel assignments (assigning channels to a cell
they do not have a call in progress. Forward control which is not adjacent in frequency).
channels also serve as beacons which continually broad- Inter-symbol interference (ISI) When the signal travls
cast all of the traffic requests for all mobiles in the system. through a channel, objects in ansmission path can create
Supervisory and data messages are sent in a number of multiple echoes of the signal. hoes occur at the receiver
ways to facilitate automatic channel changes and handoff and overlap in successive e slots. This is known as inter-
instructions for the mobiles before and during a call. symbol interference equalizers at the receiver can be used
to compensate the effect of ISI created by multi-path with-
Cellular systems rely on the frequency reuse concept, in time dispersive channels.
which requires that the forward control channels (FCCs) Fading one consequence of transmitting a signal
in neighboring cells be different. By defining a relatively through a time –varying multi-path channel is to confront
small number of FCCs as part of the common air inter- at the receiver with a signal fading (amplitude variations
face, cellular phones can be manufactured by many com- in the received signal). Hence not only the propagation
 27

delays but also the random impulse responses of the the receiver amplifiers, & inter-cellular interference. In
channel will provoke some attenuation and time spread addition, this noise can also be generated internally to the
of the signal transmitted. communications system as a result of Inter-Symbol Inter-
Thermal noise finally, the additive thermal noise is a ference, Inter-Carrier Interference & Inter-Modulation
factor that always corrupts a transmitted signal through a Distortion. These sources of noise decrease the Signal to
communication channel. Generally this thermal noise is Noise Ratio(SNR) & thus limiting the spectral efficiency
assumed to be an additive white Gaussian noise of the system. Noise is the main detrimental effect in most
(AWGN). radio communication systems.
Most types of noise present in radio communication
systems can be modelled accurately using Additive White
3 CHANNEL FADING TECHNIQUES Gaussian Noise(AWGN). This noise has a uniform spec-
The physical medium between the transmitter and tral density & a Gaussian distribution in amplitude.
reeiver is known as channel. This channel results in ran- Thermal & electrical noise from amplification, primarily
dom delay (random phase shift) with total a factor. have white Gaussian noise properties, allowing them to
Channels may be three types: be modeled accurately with AWGN. Also most other
noise sources have AWGN properties due to the trans-
Type Description Examples mission being OFDM. OFDM signals have a flat spectral
Simplex One way only FM radio, tele- density & a Gaussian amplitude distribution provided
vision that the number of carriers is large, because of this the
Half duplex Two way, only Poice Radio inter-cellular interference from other OFDM systems have
one at a time AWGN properties. For the same reason ICI, ISI, & IMD
Full duplex Two way, both Mobile systems also have AWGN properties for OFDM signals.
at the same time

3.1 Additive White Gaussain Noise

Zero-mean white Gaussian Noise (WGN) has the same
power spectral density AWGN(f) for all frequencies. The
adjective ‘white’ is used in the sense that white 3.2 Rayleigh Fading Channels
When information is transmitted in an environment
with obstacles (Non Line-of-sight - NLOS), more than one
transmission paths will appear as result of the reflec-
light contains equal amounts of all frequencies within tion(s). The receiver will then have to process a signal
the visible band of electromagnetic radiation. The auto- which is a superposition of several different transmission
correlation function of WGN is given by the inverse paths. If there exists a large number of transmission paths
Fourier transform of the noise power spectral density may be modeled as statistically independent; the central
GWGN(f): limit theorem will give the channel the statistical charac-
The autocorrelation function RaWGN (t) is zero for t≠0. teristics of a Rayleigh Distribution [36]. (Fig 5)
This means that any two different samples of WGN, no
matter how close together in time they are taken, are un-
correlated. The noise signal WGN (t) is totally decorre-
lated from its time shifted version for any t≠0.

Fig. 5 Rayleigh distribution

3.3 Multipath Fading

Fig. 4 Signal with AWGN Noise

In a radio propagation channel, there are two types of

The amplitude of ‘integrated’ (bandwidth) WGN has a fading; large scale and small scale. Large signal fading is
Gaussian probability density distribution P(WGNi): the attenuation caused by the path loss over large dis-
tances and shadowing effects and is well represented by
Noise exists in all communications systems operating log-normal models. Small scale fading occurs in the range
over an analog physical channel, such as radio. The main of the signal wavelength and is much more random as
sources are thermal background noise, electrical noise in compared to large scale fading. Small scale fading is
 28

3.6 Small Scale Fading

mainly a result of multiple multipath components under- Small scale fading itself can be subdivided into two
going constructive/destructive interference at any point. other types. One is based on the multipath delay spread
In the common case where there is no line of sight (LOS) and the other on the Doppler shift. Depending on the rel-
component between the transmitter and the receiver, the ative length of the multipath delay spread ( ) with
received power follows a Rayleigh distribution and can respect to the OFDM symbol length we can have either
be seen to have a Rayleigh fading envelope. Flat Fading or Frequency Selective Fading. Similarly, the
relative magnitude of the channel coherence time (as a
result of the Doppler shift) with respect to the OFDM
symbol duration determines whether the signal under-
goes Fast Fading or Slow Fading [10].

3.7 Flat Fading

Flat fading is the name given to the case when the chan-
nel coherence bandwidth is larger than the signal band-
width and hence all frequencies of the transmitted signal
Fig. 6 Typical Rayleigh fading envelope at 1 kHz Doppler spread
experience the same channel condition; i.e., over the sig-
(0.423ms coherence time) nal bandwidth, the channel frequency response is essen-
tially flat; and hence the name Flat Fading. In the time
domain, this corresponds to having an expected
3.4 Large Scale Fading smaller than the signal symbol period.
The large scale fading refers to the degradation of the 3.8 Frequency Selective Fading
signal strength due to the path loss as a function of the
On the other hand, if the channel bandwidth is nar-
distance between the transmitter and the receiver and
rower than the signal bandwidth, different frequency
shadowing effects caused by the surrounding environ-
bands of the signal are affected differently. The time do-
mental clutter. The path loss is the gradual loss of re-
main analogue is that the channel is larger than the
ceived signal power with the distance from the transmit-
signal symbol period.
ter. Shadowing describes the random effects which occur
at different locations which have the same transmitter 3.9 Fast Fading
receiver separation, but different surroundings on the In a fast fading channel, the rate of change of the channel
propagation path. is higher than the signal symbol period and hence the
channel changes over one period. In other words, the
3.5 Shadowing channel coherence time, Tc, is smaller than the symbol
period. Tc is related to the Doppler spread, fm, as
From the above equation only gives an ensemble aver-
age value, which does not take into account the variability
Tc = 0.423/fm
in the propagation path between the transmitter and the
From this relation it is clear that a high Doppler spread
receiver. Empirical data measurements have shown that
results in a smaller channel coherence time. The cohe-
the received signal power at a particular distance d is dis-
rence time of 0.423ms corresponding to a fm of 1kHz is
tributed log-normally (normally in dB) about the dis-
clear.
tance-dependent average received signal power. When
the log-normal shadowing is included, the large scale 3.10 Slow Fading
fading at a distance d, can be expressed as:
As the name suggests, in a slow fading channel, the
Where X represents a normally distributed random va- channel coherence time is larger than the symbol period
riable with zero mean and variance . The variance of and hence the channel remains approximately static over
the log-normal shadowing is calculated based on mea- a symbol or multiple symbols. From the above equation it
surements that are taken over a wide range of locations is clear that slow fading is usually expected with low
that have the same transmitter receiver separation but Doppler spread values; i.e. with slower moving obstacles
with different environmental clutter. Apart from this and receiver/transmitter. Multipath delay spread based
simple model of the above equation, there exist various and Doppler spread based fades are completely indepen-
empirical models that take into account the terrain profile dent of each other and hence is quite possible to have a
of a particular area in the propagation predictions (such flat, fast fading channel or a flat, slow fading channel; and
as Okumura model [11] and Hata model [12]). so on.
 29

4 RESULT AND DISCUSSION C. BER of image vs random data

I discuss my simulation result by two steps. First I ana- The correlation between image pixels does not effect
lyze the performance comparison by different parameter the BER in AWGN channel.
setting in each channel. Then I analyze the performance 4.2 For Flat Fading Channel
by comparing three different channels under the same
parameters setting. All the simulations are based on A. BER of simulation vs theoretical
QPSK modulation with gray code. As shown in figure 9 the BER performance of simula-
4.1 For AWGN channel tion result is worse than theoretical BER. This is reasona-
ble, since the theoretical BER is based on the assumption
A. BER of simulation vs theoretical that we know exactly the phase information of modulated
As shown in figure 7, The BER performance of imula- signal. However, due to the time-variant channel, we al-
tion result is closely identical to theoretical BER. ways have estimation error for phase information. We
also find the BER performance is improved dramatically
in low SNR, while not in high SNR. This is also reasona-
ble, since in low SNR, white Gaussian noise dominate the
BER error, which can be improved by enhancing SNR,
while in high SNR, phase estimation error dominate the
BER error, which can not be improved by simply enhanc-
ing SNR .

Fig. 7 BER of Simulation vs Theoretical

B. Image quality of received vs original

In figure 8, the received image is plot at SNR = 5dB, we
see there are some random noises in the image. From si-
mulation result, the received image quality is almost the
same as original at SNR = 10Db.

Fig. 9 BER of simulation vs theoretical

B. BER & constellation of training vs non-training

As shown in figure 11 and figure 12 the constellation is
plot at SNR = 10dB, we see both the BER performance
and constellation are greatly improved by channel phase
estimation.

Fig. 8(a) Original

Fig. 11 BER & Constellation of training vs nontraining

Fig. 8(a) Image quality of received

 30

(a) Fig. 13 BER of simulation vs theoretical

B. BER & constellation equalized vs non-equalized

In figure 13 and figure 14 the constellation is plot at
SNR = 15dB, we see both the BER performance and con-
stellation are greatly improved by channel phase estima-
tion.

(b)

Fig. 12(a) Without adjustment (b) With adjustment

C. Image quality of received vs adjusted

In figure 6, the received image is plot at SNR = 10dB,
we see that other than some random noise, there is some
block noise in the image. This is due to the phase estima- Fig. 14 BER & Constellation of equalized vs nonequal
tion error in a coherence time.

D. BER of image vs random data C. Reset vs continue training result

The correlation between image pixels does not affect the The BER performances of resetting the state of equaliz-
BER in flat fading channel. er come from training result of last coherence time is
worse than using the result of last coherence time. The
4.3 For frequency selective fading channel
BER is improved by using decision directed mode, since
A. BER of simulation vs theoretical the time-variant property of the channel cause the chan-
nel change from estimation result of training data.
As shown in figure 13, the BER performance of simula-
tion result is worse than theoretical BER. The reason is D. Training only vs decision directed mode
same from above reason addressed in flat fading channel. The BER is improved by using decision directed mode,
Different from in flat fading channel, the BER perfor- since the time-variant property of the channel cause the
mance is improved dramatically in low SNR, while even channel change from estimation result of training data.
degraded in high SNR. This is also reasonable, since in
high SNR, phase estimation error and ISI dominate the E. Training only vs decision directed mode
BER error, and the estimation error will cause even severe The BER is improved by using decision directed mode,
ISI., which cause the BER even worse. since the time-variant property of the channel cause the
channel change from estimation result of training data.
 31

F. LMS vs RLS
The BER performances are almost same for both of
them. But during the simulation, we find, LMS need more
training data to converge the equalizer comparing to RLS,
while latter has more complexity and time consuming.

G. Image quality of received vs original

In figure 15 the received image is plot at SNR = 15dB,
we see that other than some random noise and block
noise in the image, there are some overlaps in the image.
This is due to the white Gaussian noise, phase estimation
error in a coherence time, and ISI caused by frequency
selective fading channel.

(a)

(a) (b)

(b)
Fig. 15 (a) without equalization (b) with equalization (c)

H. BER of Image vs random data

The correlation between image pixels does not affect
the BER in frequency selective fading channel, since we
use PN code to train the equalizer.

4.4 Comparison among three channels

A. For Image comparison

In figure 16, 17, we may see that in AWGN channel,
the image is degraded by random noise; in flat fading
channel, the image is degraded by random noise and (d)
block noise; in frequency selective fading channel, the
image is degraded by random noise Fig. 16 (a) Original image (b) AWGN channel (c)
Flat fading channel (d) Frequency selective fading chan-
nel
 32

Channel’, IEEE Trans., Aug 1974, IT-22, pp. 1064-1075

[6] Ziv, J.: ‘Probability of decoding error for random pha-
seand Rayleigh fading channels’, IEEE Trans., Jan 1965, IT-
11, pp. 53- 61
[7] M. Pukkila, “Channel Estimation Modeling”, 2000
[8] CROZIER, S., FALCONER, D., and MAHMOUD,
S.:‘Shortblock equalization techniques employing channel
estimation for fading timedispersive channels’. IEEE Vehi-
cular Technology Conference, May 1989, pp.142-146
[9] UNGERBOECK, G.: ‘Adaptive maximum-likelihood re-
ceiver for carrier-modulated data-transmission systems
[10] Do-Sik Yoo; Hafeez, A.; Stark, W.E., “Trellis-based multiuser
detection for DS-CDMA systems with frequency-selective
fading, ” Wireless Communications and Networking Con-
ference, IEEE, 1999, vol.2. pp. 829 –833.
Fig. 17 (a) AWGN channel (b) Flat fading channel (c) Frequency [11] J. G. Proakis, Digital Communication, 3rd edition,McGraw-
selective fading channel hill,1995.
[12] W. C. Jakes, “Microwave mobile communication,” IEEE
deep spectral nulls in the passband. While frequency se- Press Classis Reisse,1974.
lective fading channel normally causes the deep spectral
nulls, so in our future simulation, we may improve this
Biography
by add Decision Feedback Equalization (DFE). In this
project, we produce two different scenarios by simulate a
GSM carrier frequency and bandwidth, and use pilot data Md. Sipon Miah received the Bachelor’s
to estimate the channel phase. All of these are simulated and Master‘s Degree in Information and
in MATLAB at present. In our future model, we may in- Communication Engineering from Islam-
tegrate our model into GNU radio with USRP hardware ic University, Kushtia, in 2006 and 2008,
respectively. He is courrently Lecturer in
support, which will give a practical environment to test the department of ICE, Islamic Universi-
our wireless communications simulation and our own ty, Kushtia-7003, Bangladesh. Since
algorithm. 2003, he has been a Research Scientist
at the Communication Reasearch La-
boratory, Department of ICE, Islamic
5. CONCLUSION University, Kushtia, where he belongs to the spread-spectrum re-
search group. He is pursuing research in the area of internetworking
In this paper, I test the effect of three different channel in wireless communication. He has one published paper in interna-
models, AWGN channel, flat fading channel, and fre- tional and national journals in the same area but under process. His
are as of interest. Include database system, optical fiber commnica-
quency selective fading channel, to the data and image tion and mobile communication.
under two scenarios. I also compare and analysis the im-
provement of channel estimation and adaptive equaliza-
tion in slow fading channel. Our result is exactly identical M. Mahbubur Rahman received the
Bachelor’s and Master‘s Degree in Phys-
to the theoretical analysis. I also propose some possible ics, Rajshahi University, in 1983, 1994
improvements and probable future work, which will in- and PhD degree in Computer Science &
troduce more research interests. Engineering in 1997. He is courrently
Professor in the department of ICE, Is-
lamic University, Kushtia-7003, Bangla-
ACKNOWLEDGMENT desh. He has twenty four published pa-
pers in international and national jour-
The The authors would like to thank student Md. Far- nals. His are as of interest include inter-
zan Ali [M.Sc.(Tech.)] for her assistance during the com- networking, AI & mobile communication.
puter simulations and Mst. Tania Parvin [M.Sc. (Mat.)] for
the language checking. Tapan Kumar Godder received the
Bachelor’s, Master’s and MPh degree in
Applied Physics & Electronics from Raj-
REFERENCES shahi University, Rajshahi. In 1994, 1995
[1] T. S. Rappaport, “Wireless Communications: Principles and 2007, respectively. He is courrently
Associate Professor in the department of
and Practice”, Second Edition, 2002 ICE, Islamic University, Kushtia-7003,
[2] J. G. Proakis, “Digital Communications”, Fourth Edition, Bangladesh. He has seventeen pub-
2001 lished papers in international and na-
[3] S. Haykin, “Adaptive Filter Theory”, Fourth Edition, 2002 tional journals. His are as of interest
include internetworking, AI & mobile communication.
[4] A. V. Oppenheim, R. W. Schafer, J. R. Buck, “Discrete time
Signal Processing”, Second Edition, 1999
[5] Monsen, P.: ‘Adaptive Equalization of the Slow Fading
 33

Bikash Chandran Singh received the

Bachelor’s and Master‘s Degree in Infor-
mation and Communication Engineering
from Islamic University, Kushtia, in 2005
and 2007, respectively. He is courrently
Lecturer in the department of ICE, Islamic
University, Kushtia-7003, Bangladesh. His
are as of interest including database, opti-
cal fiber communication & multimedia.