SISO Thesis

Performance analysis of orthogonal frequency division multiplexing (OFDM) using different
modulation techniques for image encryption
CHAPTER -1
INTRODUCTION
DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 1

1.1 INTRODUCTION
This chapter includes the motivation, problem definition, approaches and scope of the
thesis. It describes the basic theme of the thesis and provides overall idea about thesis. In recent
years we have witnessed an increasing popularity of multimedia applications that run on personal
mobile devices. Services such as high quality video calls, mobile TV, audio and video contents
on demand and various interactive map/locator services (such as GPS) are becoming widely
supported in new generations of personal mobile devices. A majority of these services require a
minimal guaranteed data rate between users (or between a user and a base station) in order to
provide a minimal predefined quality of service, which puts a demand on transmission
bandwidth and need for spectral efficiency on wireless channels. To provide support for
bandwidth demanding applications on the physical layer, future standards need to specify both
bandwidth requirements and type of signaling that achieves the data rate required for minimal
predefined qualities of services for future applications.
The third generation (3G) mobile communication systems have become popular all
around the world. But, its services cannot provide a very big dynamic range of data rates, nor can
it meet the requirements of a variety of business types. Besides, voice transportation in 3G still
relies on circuit switching technology, which is the same method as used in second-generation
(2G) communication systems, rather than pure Internet Protocol (IP) approach. Thus, based on
consideration listed above, many countries have already carried out research on the next
completely evolutionary fourth generation (4G) communication systems which provide a
comprehensive and secure IP solution where voice, data, and multimedia can be offered to users
at anytime, wherever the user wants with higher data rates than previous generations [1]. Since
bandwidth resource in 4G mobile communications [2] is still scarce, in order to improve
spectrum efficiency [3] and achieve as high as 100Mbps wireless transmission rate, it requires
more advanced techniques to be employed. The limitation of modulation schemes in existing
communication systems has become an obstruction in further increasing the data rate. Hence,
next generation mobile communication systems need more sophisticated modula tion scheme
and information transmission structure.

Orthogonal frequency division multiplexing (OFDM) have therefore been adopted due to
their superior performance. They promise to become key high-speed wireless communication
technologies [4] and combining them can provide wireless industry evolution from 3G to 4G
system.
1.2 Overview
Simulation of wireless channels accurately is very important for the design and
performance evaluation of wireless communication systems and components. Fading or loss of
signals is a very important phenomenon and must be well understood by all engineers related to
the Wireless Communications Field. That leads us to the fading models which try to describe the
fading patterns in different environments and conditions. Although no model can perfectly
describe an environment, they strive to obtain as much precision as possible. The better a model
can describe a fading environment, the better can it be compensated with other signals, so that,
on the receiving end, the signal is error free or at least close to being error free? This would mean
higher clarity of voice and higher accuracy of data transmitted over wireless medium.
Frequency bands used by mobile devices are strictly specified by responsible regulatory
bodies, which set limits on the bandwidth available for communication. Therefore, a very natural
and important question is what the maximum data rate is (equivalently, information rate) at
which reliable communication over a mobile channel of a given bandwidth is attainable. This
quantity is known as the channel capacity.
For AWGN channels of a given bandwidth, Shannon [1] has derived the well-known
expression for the maximum data rate that can be achieved, for reliable communication. That is,
the average bit error rate (BER) can be made arbitrarily close to zero by use of channel coding,
for transmissions up to the maximum achievable rate. For mobile channels, that are time-varying
and dispersive in time and frequency however, the channel capacity derivation is still an open
research area. In this context, we point out the lack of equivalent vector channel models for
realistic continuous-time SISO dispersive fading channels. Such models serve as the foundation
upon which channel capacity results are derived.

The study of the achievable average error rate performance of communication systems is
a natural complement to the study of channel capacity, since waveform fading channel models or
their equivalent vector channel models are now used in the context of actual communication
systems which include a transmitter and receiver. The average error rate performance, quantified
via the minimal average symbol error probability, is a measure of communication reliability.
Given a realistic fading channel model and a transmitter configuration, the study objective is to
minimize the error rate performance by an appropriate choice of receiver design.
An example of a common signaling scheme and a common received-signal
discrimination technique is the use of Nyquist sampling of the received signal. This approach is
neither optimal in the sense of minimizing the average error rate performance, nor have they
been previously compared (in sense of the achievable error performance) to optimal
communication systems that perform an ML detection of the received data symbols. Nyquist
signal sampling doesn't maximize the SNR in the received signal observables when the received
signal is not band limited. Additionally, it can be information-lossy when the number of samples
is finite [2]. Also, the interference cancellation techniques, commonly used prior to symbol
detection, diminish the positive effects of implicit channel diversity [3] ordered by
time/frequency-selective fading channels.
1.3 Objective and Motivation for current Study
With the evolution of the wireless system the demand for high speed data services have
been increasing day by day, which is impossible to be achieve by the conventional serial data
transmission system without trade-off between high speed data services and QOS without
increasing the band width of the system. Here both the options are inconvenient, as one never
demands the degradation of the service quality ( because if we increase data rate in serial data
transmission ISI will gradually increase which make the extraction of actual information at
receiver nearly impossible ) and secondly the need for extra spectrum in a limited spectrum
scenario. In order to overcome this problem new parallel data transmission system was proposed,
which is known as OFDM system. Here AWGN, Rayleigh channels have been used for analysis
purpose and their effect on BER and Spectral Efficiency for high data rates have been presented.
The proposed system has better performance than the other systems in terms of SNR

improvement and spectral efficiency. Here Alamouti 2 X NR Space Time Coding is used for
analysis purpose. It is shown that the using antennas at transmitter or receiver side provides
significant gains over single transmit single receive antenna (SISO) systems.
Wireless communication standards which arein use and under development, adopt
orconsider adopting orthogonal frequency-division multiplexing (OFDM) as new modulation
technique. OFDM has become definitive modulationscheme in future wireless communication
systems.For better ways of living has been instrumental in advancing human
civilization.Communication services are available at any time and place free people from the
limitation ofbeing attached to certain devices. Nowadays, thanks to the progress in
wirelesstechnology, affordable wireless communication service has become a reality. Mobile
phonesmade people comfortable whenever and wherever they want. Digital audio and video
broadcastinggives consumers high-resolution, better-quality and interactive programs.
Thedevices are thin, light, small and inexpensive in their form.smart mobile phones arecapable
of multimedia and broadband internet access are being shown up on the shelves.
Many projects studying wireless networks with variable extents of coverage are in the
processes. They will make wireless access to internet backbone from, either indoors oroutdoors
and in rural or metropolitan areas. In this, their evolution and futureprogress will be seen. The
important role that the orthogonal frequency-divisionmultiplexing (OFDM) technique used in
wireless communication systems will become every clear.Now, most people fulfill the need for
information and entertainment through audioand video broadcasting.
The inauguration of AM radio can be seen to the early twentiethcentury, whilst analog
TV programs were first broadcast in the Second World War. In middle of twentieth century, FM
radio programs were available. Thesetechnology, were on analog communication, brought news,
music, drama, movies andmuch more into our lives. To provide more programs, digital
broadcastingtechniques, such as digital audio broadcasting (DAB) and digital video broadcasting
(DVB), started to substitute the analog broadcasting technologies in the past several years.
DAB is among the initial standards that use the OFDM technique. The DAB project
began inmid-1980 on OFDM, DAB has one benefit: a single-frequency network(SFN). In a
single frequency broadcasting network, one carrier frequency can be taken for alltransmitters to

broadcast the same radio program in the country without co-channel interference. But in the FM
system, one out ofapproximately 15 possible frequencies can be utilized, resulting in ainefficient
range of frequencyreuse factor of 15. A single-frequency network and a multi-frequency
network.
In the DAB system, it is not important to go for radio stations as withAM/FM radios. The
programs of all radio stations are combined in multiplexes.Multiplexes save maintenance cost
of radio stations. On the other hand, variablebandwidths can be provided to each programmer,
fulfilling their related demandsfor good quality. Music radio multiplexes can be transmitted at a
rate which is up to the highest-quality192 Kbps, on the other hand mono talk and news programs
use only 80 Kbps. Moreover, theDAB system features better mobile receiving quality by OFDM
technique.
It is European standard for digital television broadcasting. DVB standardswill
have(DVB-S) for satellites, DVB-C for cables, DVB-T for terrestrial transmission andDVB-H
for low-power handheld terminals. In them, DVB-T and DVB-H use OFDMas the modulation
scheme. DVB-T receiver initial shipping in late-1990 and now digitalDVB-T programs are
available in different countries. As the DAB system, DVB-T/Htechnology also takesingle-
frequency networkscountry wide. In additionto this DVB-T/Hstandards give several modes of
operation that are used for large-scale SFN and high mobilityreceiving.
The core digital stream in DVB-T will be MPEG-2 transport stream that have one or
more program streams. Each and every stream of multiplexing compressed video, audio and
datasignals. The DVB-T standard can take a data rate of MPEG-2 high-definition TV(HDTV), it
is up to 31 Mbps. In DVB-H, high-speed IP services as an improvementof mobile
telecommunication networks are being provided. , DVB standard has been allowed for
integration with bi-direction data connections with other access technology, thusmaking
interactive applications between the viewers and the TV stations.
Phones are necessity to several billions of people in the globe. Working range from voice
service to video and broadband data service and many more.The transformation from the second-
generation,(2G) to the third-generation(3G), and then again to the fourth-generation (4G) mobile
cellular communicationsystems. In 2G, the GSM systems are used as the European standard and

CDMAOne IS-95is taken in North America. Each of them provideddigital voice services at 10
Kbps.Afterwards, General Packet Radio Service and Enhanced Data rate for GlobalEvolution
(EDGE) systems offers transmission rates of up to several hundreds of Kbps asan improvement
of the GSM standard. Same as in, CDMA2000 1X modified the datatransmission to 300 Kbps in
North America.
Now a day, 3rdGen.Standardprovides data services with a data rate of 2 Mbps tocombine
multimedia applications. Two other main-stream 3G standards which is in use are CDMA2000
3Xand wideband-CDMA (W-CDMA). The modified version of W-CDMA has been
standardizedas High Speed Downlink Packet Access (HSDPA), which is assumed as 3.5G
andcan bring about a 10-Mbps transmission rate. The third-Generation Partnership
Project(3GPP) long-term evolution (LTE) has begun to plan real solutions to upcoming
mobilecommunication technology. The main features include
Spectral efficiency up to10 b/s/Hz;
Provision of a radio resource management for increasing cell coverage and
modifiedSystem efficiency;
To support internet protocol version 6 (IPv6) multimedia services with low
powerconsumption and high performance.
To support mobility up to 250 Km/hr.
For satisfying high spectral efficiency, less power consumption and bestperformance need,
modern techniques are important in near future 4G system.
Modern wireless communication systems give many advantages, such as mobility, easy
access to signal and device installation; however, they have more limitations than wireline
transmission systems it has limited capacity, spectrum shortage and some service quality
uncertainties. With wireless communication systems, signals are sent over the air in the radio-
frequency (RF) band.
There are many non-ideal factors that change the quality of the receiving signals and, thus,
the reliability of modern wireless communication. As an outcome, these wireless channel
changes places of different fundamental limitation on the capability of modernwireless
communication systems.

Our wireless radio channels are dynamic and time-varying due to channel variations and user
mobility. Leading to matters worse is the fact that wireless transmission is off and on due to the
irregular nature of the wireless channels. Therefore, most of the analysis and estimation of the
modern wireless channels must be started. In this the effects of the modern wireless channels, the
transmission signal also has distortions. It is caused by RF front-end circuits. To design an
equivalent baseband channel model for a modern wireless communication system, signal
propagation characteristics as well as different circuit improvements should be taken into our
knowledge. Wireless baseband channel modeling is very decisive to the design of wireless
baseband receivers, if any inaccuracy inevitably in it causes either deficient receiver designs or
its costly receivers that are over-designed.
Modern wireless communication system, because of many hurdles in the propagation, RF
signals usually move along with different paths, give rise to reflection, scattering and also for
diffraction. The combined effect of the above mechanisms is irregular and intricate. Therefore,
these different propagation processes are combined and subdivided as path loss, shadowing and
multipath fading.
1.4 Ageof Modern Wireless Communications
Modern wireless communications is, the fastest growing part of the communications
industry. ithas gathered the attention of the media and the imagination of the people. Cellular
mobile systems have seen outstanding progress over the last decade and there are now two
billion users in the globe. Really, mobile phones have become a important trading tool and part
of our life in most developed countries, and are supplanting antiquated wire line systems in
developing countries. Modern wireless local area networks now replace wired networks in
different houses, businesses, and campuses.
Many modern applications, which have wireless sensor networks, automatic highways
and industries, smart homes and appliances, and distanced telemedicine, are from research ideas
to solid systems. Past of wireless communications the initial wireless networks were developed
before -industrial age. These systems sent information over line-of-sight distances (later
improved by telescopes) taking smoke signals, torch signaling, flashing mirrors, signal flares, or
semaphore flags etc. A set of signal combinations were developed to send complex messages

along rudimentary form of signals. Different observation stations were set up on hilltops and
along the roads to relay these messages over very large distances. These new communication
networks were substituted first by the telegraph network (invented by Samuel Morse in 1838)
and later by the telephone. After 1895, in a few decades the telephones were invented,
The great scientist Marconi demonstrated the first radio transmission from the Isle of
Wight to a tugboat which was 18 miles away, and radio communications was developed The
initial network depend on packet radio, ALOHANET, was developed at the University of Hawaii
in the year 1971. Network made computer sites at seven campuses extended to four islands for
communication with a central computer on Oahu via radio transmission.
Architecture of network used a star topology with the central computer at a hub. In
American U.S. military was extremely interested in the combination of packet data and broadcast
radio inherent to ALOHANET. Throughout theyear 1970s and early 1980s the Defense
Advanced Research Projects Agency (DARPA) spend resources to create networks using packet
radios for tactical communications in the field. Packet radio networks seen commercial
application in having wide-area wireless data services first initiated in the early 1990s, enable
modern wireless data access at low speeds, order of 20 Kbps. market for these wide-area
wireless data services never being materialized, due to their low data rates, high cost, and lack of
their bad applications. Servicesdisappeared in the 1990s, supplanted by the wireless data
capabilities of cellular telephones and wireless local area networks with the advent of wired
Ethernet technology in the year 1970s steered many commercial companies out from radio-
based networking.
In the year 1985 the Federal Communications Commission made the commercial
development of wirelessLANs by authorizing the public use of the Industrial, Scientific, and
Medical frequency bands for wirelessLAN products. The ISM bandswere very attractive to
wireless LAN vendors since then they did not need to get an FCClicense to operate on this band.
Moreover, the interference from first users within this frequency band was very high result of
this initial wireless LANs had bad performance in terms of data rates and its coverage.
The exiting generation of wireless LANs, based upon the family of IEEE 802.11
standards. Wired Ethernets today give data rates of 100 Mbps and the performance gap

betweenwired and wireless LANs will increase with time without additional spectrum allocation.
Apart from the big data rate differences, wireless LANs is becoming the attractive Internet
access method in different houses offices, and campus due to their convenience and freedom
from wires. However, most of the modern wireless LAN support applications such as email and
web browsing that are not now bandwidth-intensive.The most successful application of wireless
networking has been the cellular telephone system so far.
The roots of this system started in the year 1915, when wireless voice transmission
between San Francisco and NewYork was first set up. In the year 1946 public mobile telephone
service was adopted in 25 different cities in entire United States: 30 years after the introduction
of mobile telephone service the New York system had to support 543 users. Asolution of
capacity problem emerged during the 50s and 60s when engineers and researchers at AT&T
Bell Laboratoriesdeveloped the mobile and cellular concept. In the year 1947 AT&T asked
spectrum formobile and cellular service from the FCC. The design was done bythe end of the
year 1960s, the first field test wasseen inthe year 1978, and the FCC granted servicesystem
authorizationin the year 1982.
1.5.Modern Day Wireless Systems

It has been seen that subcarrier spacing in OFDM systems must be carefully designed so
that each subcarrier suffers flat fading sometimes, the extent of channel coherence find the
maximum subcarrier spacing. From the other side due to parallel transmission on numerous
subcarriers, the OFDM symbol period is adopted longer than that of the single-carrier
symbols.Badly, time-selective fading takes the long OFDM symbols. While in case in which the
channel acknowledgement varies significantly within an OFDM symbol, AWGN will come up
with, degrading the system performance.
The OFDM symbol period has been upper bounded by the length of the interval that the
channel remain stable symbol period is the reverse of the subcarrier spacing, the minimum
subcarrier spacing is thus found by how fast the channel is vary. Assigned the signal bandwidth
and the subcarrier spacing, the FFT size can then be determined. The guard interval depends on
the channel maximum delay, and can be adjusted on the operating environment. The guard band

ratio has been adopted the sidelobe attenuation to fit in the spectrum mask. The data transmission
rates can be calculated by using the symbol period, the number of subcarriers and the adopted
constellation size, taking into consideration the overhead of guard interval and guard
band.Digital modulation has been the process of representing binary information using segments
of several sinusoidal waveforms. The parameters and calculations that can be modified in a
sinusoidal wave are its amplitude, frequency and phase.
The digital audio broadcasting (DAB) system has four operation modes for varying
transmission bands using different FFT sizes, now the same 1.5-MHz bandwidth. As we take
higher the carrier frequency, the wider the subcarrier spacing and the shorter the symbol period
so that it becomes able to resist time selective fading. The same guard interval ratio can be used
for all four modes. Hence, the mode is good for operating in the long-delay-spread channels,
namely large-scale cells. A quarter of subcarriers are reserved with respect to guard band.
The digital video broadcasting-terrestrial (DVB-T) system gives out digital video signals
in the frequency band ranging from 400 to 800 MHz Each DVB-T signal can be combined in a
channel with either of 6,7 or 8-MHz bandwidth. DVB-T works with 2K FFT and 8K FFT. In the
2K mode, Total 1705 out of 2048 subcarriers can be used to carry the data and pilot signals. As
the symbol period is get shorted, this mode can be designed to deal with the scenarios with fast-
changing channel conditions. From the other part, the 8K mode has a longer guard interval which
is making it capable of coping with multipath channels with a long delay spread. Total 8192
subcarriers in the 8K-mode DVB-T signal, 6817 can be used for data and pilot transmission.
Now with the high-order 64-QAM constellation, the highest data rate of the 8-MHz
DVB-T system will be 31.67 Mbps with a code rate of 7/8.
The IEEE 802.11a/g wireless LAN can work in ISM band around 2.42.5 GHz and in U-
NII band around 5.8 GHz A total of 64 subcarriers will divide 20-MHz bandwidth. This high-
speed wireless LAN standard makes stationary reception in indoor environments. Our system
will have the guard band ratio and the guard interval ratio 0.1875 and 0.25, respectively, and the
maximum information data rate will be around 54 Mbps.
The wireless MAN, IEEE 802.16e-2005, having three modes: single-carrier, OFDM and
OFDMA .In the OFDM mode, an FFT size of 256 can be used for different bandwidths ranging

from 1.5 to 28 MHz and the carrier frequency can be from 2 to 11 GHz. The sampling frequency
can be 8/7 or 7/6 times the bandwidth and the guard interval ratio changes from 1/32 to 1/4. It is
protected by a code with rate 3/4, the highest data rate will be 104.7 Mbps, which will have the
smallest guard interval overhead and the widest channel bandwidth.
The 3GPP-long-term evolution (3GPP-LTE) has evolved universal terrestrial radio access
(EUTRA)project, willing to improve the UMT mobile phone standard to fulfil fourth-generation
(4G) requirementsit has adopted OFDMA with different FFT sizes in the downlink transmission.
Its bandwidth changes from the range of 1.25 to 20 MHz for giving flexible allocation. The
subcarrier spacing can be taken at 15 KHz, which as an output in possible FFT sizes ranging
from 128 to 2048. The maximum cell size can be 5 Km so that 16.67 ms can be reserved for the
guard interval. The target downlink data rate is likely at 100 Mbps in the 20-MHz bandwidth.
1.6 Modes of Motivation
In frequency division multiplexing with image processing, the total bandwidth can be
divided into number of number of overlappingfrequency with sub channels. Then sub channelis
modulatedor modified with differentsymbol and then it is multiplexed. There is a guard band
between each sub channels, soas to avoid peak average power ratio &inter symbol interference
with AWGN. This resulted to inefficient way of utilizing theexistingbandwidth
In [3] was proposed an idea to deal with wastefulness ofspectrum by development of
overlapping sub channels in frequency division multiplexing for image to avoid AWGN, these
sub channels must be orthogonal to eachother.
OFDM has proposed as the core technique for the fourth generation (4G)
wirelesscommunication system. Currently, OFDM is being used in many wireless
communicationsystems, such as Wireless Local Area Network (WLAN) systems,
HIPERLAN2(High Performance Local Area Network), Digital Video Broadcasting (DVB)
systems,Worldwide Interoperability for Microwave Access (WiMAX).
The main limitation of OFDM system is PAPR. The PAPR is caused due to orthogonally
loss between the sub carriers. The orthogonallyloss may be caused by phase noise,timing offset,
carrier frequency offset (CFO) and Doppler frequency shift and AWGN. The PAPR induced by
the first important impairments can be completely cancelled, but due to randomnessof average

power, it can only be mitigate.

1.7Structure of Thesis
The thesis organization is as follows.
Chapter 1: Introduction of OFDM.in which the functioning and basic understanding is discussed
Chapter 2: Literature survey and problem formulation.
Chapter 3: Basics of OFDM system and its block diagram is presented. Function of IFFT and FT
is also discussed. How guard time and cycle extension removed PAPR and ICI.
Chapter 4: This chapter explains about OFDM transmitter and receiver.In this image is Sent over
the system and AWGN is removed.
Chapter 5: Simulation and Resultexplain the working about the original image
throughMATLAB.
Chapter 6: Future work, application of OFDM and conclusion is presented.

CHAPTER 2
LITERATURE SURVEY

In this chapter we have provided the background necessary for the rest of the dissertation.
It reviews related concepts in the context of the work presented in this dissertation. Main
contribution of the research presented in this dissertation is to establish architecture for several
applications; we then focus our review on several novel technologies that are adopted by our
architecture to address interaction and the related issues are discussed
2.1 Literature Survey
In [1] Patrick J. Langfeld, at.al. (2012). The different sources of errors in synchronising
an OFDM system have been proposed and their impact on the demodulated information symbol
in the receiver has been investigated. As a result an analytical model of the demodulated signal
was elaborated, including all distortions caused by synchronization inaccuracy. In particular this
model allows a classification of all distortions in static and dynamic parts. Moreover it enables to
evaluate efficient methods of estimating and correcting the occurring errors separately.
In [2] H. Meng,Y. L. Guan at.al coding Spreading, etc. an easy way is clipping but it
provide degradation BER and distort the signal. Some more techniques are far better techniques
than this but they need more information, so the transmission rate comes down. on reducing
AWGN, other aspect like complexity, transmission rate, BER, error correction etc. also is taken
into consideration.
In [3] G. Avril, M. Tlich, at.al have done lots of work and researches on efficient OFDM
transmission and AWGN reduction. It is worth mentioning that understanding the impulsive
noise characteristics of electrical devices individually is essential from the communication
aspect. In this respect, some results on noise characteristics of different electrical appliances
They have proposed, some of them established various techniques to reduce the AWGN.
Other uses various schemes to create efficient sequences to minimize the AWGN to minimum
level with better error correction and reduced BER.
In [4]E.Yavuz, F. Kural, N. Coban, B. Ercan and M. Safak have performed a great work
in the field of creating best Golay complementary sequences. Nothing in above has implemented
the generated Golay sequences with MATLAB to give view of the actual performance of the
OFDM system.
In [5] C.E.Shannon.et.alThe fundamental problem of communication is that of
reproducing at one point either exactly or approximately a message selected at another point.
Frequently the messages have meaning; that is they refer to or are correlated according to some

system with certain physical or conceptual entities. These semantic aspects of communication
are irrelevant to the engineering problem. The significant aspect is that the actual message is one
selected from a set of possible messages. The system must be designed to operate for each
possible selection, not just the one which will actually be chosen since this is unknown at the
time of design. If the number of messages in the set is finite then this number or any monotonic
function of this number can be regarded as a measure of the information produced when one
message is chosen from the set, all choices being equally likely. As was pointed out by Hartley
the most natural choice is the logarithmic function. In the case of a discrete source of information
we were able to determine a definite rate of generating information, namely the entropy of the
underlying stochastic process. With a continuous source the situation is considerably more
involved. In the first place a continuously variable quantity can assume an infinite number of
values and requires, therefore, an infinite number of binary digits for exact specification.
In [6] F. R. P. Cavalcanti and S. Andersson digital communication systems has been
greatly developed in the past few years and offers a high quality of transmission in both wired
and wireless communication environments. Coupled with advances in new modulation
techniques, Orthogonal Frequency Division Multiplexing (OFDM) is a well-known digital
multicarrier communication technique and one of the best methods of digital data transmission
over a limited bandwidth.
In [7] A.R.S. Bahai, B. R. Saltzberg and M. Ergen Multicarrier Communication involves
splitting of the signal to give a number of signals over that frequency range. Each of these signals
are individually modulated and transmitted over the channel. At the receiver end, these signals
are demodulated and recombined to obtain the original signal.
J. Isabona, and M. E. Ekpenyong[8]OFDM is a method that permits the transmission of
high data rates over extremely hostile channels at a comparable complexity. OFDMs spread
spectrum technique distributes data over a large number of carriers that are spaced apart at
precise frequencies. The spacing provides the orthogonality in this technique which prevents
the demodulators from seeing frequencies other than their own. The benefits of OFDM are high
data link, resiliency to Radio Frequency (RF) interference and low multipath distortion.
Throughput is a key measure of the quality of a wireless link. It is defined as the data rate
successfully received without error per second and we would naturally prefer that this quantity
be as high as possible. OFDM is a promising technique that would satisfy the demands of the

next generation of wireless data communication. This technique will improve the quality of data
transmission link. This research studied the effect of the different parameters as they affect data
quality during transmission with respect to the channel being used. A generic throughput model
was also proposed for this technique.
In [9] M. Noh, and H. Park at.al and M. Ergen The LS estimation is the simplest channel
estimation. This algorithm has lower complexity. However, it has larger mean square error
(MSE) and easily influenced by noise and inters- carrier interference. Linear minimum mean
square error (LMMSE) algorithm.
In [10] Orthogonal Frequency Division Multiplexing (OFDM) is a multicarrier
transmission technique, which divides the bandwidth into many carriers; each one is modulated
by a low rate data stream. In term of multiple access technique, OFDM is similar to FDMA in
that the multiple user access is achieved by subdividing the available bandwidth into multiple
channels that are then allocated to users. However, OFDM uses the spectrum much more
efficiently by spacing the channels much closer together. This is achieved by making all the
carriers orthogonal to one another, preventing interference between the closely spaced carriers.
In [11] Yen.Dan Raphaeli at.al and L. Hanzo,at.al established bounds on the region of
desired triples (R, d, AWGN (C)) where R is the code rate and (d) is the minimum Eucli dean
distance of the code. They proved a lower bound on AWGN in terms of R and (d) and show that
there exist asymptotically good codes whose modulation(modification) is at most 8 log n. They
show explicit constructions of error-correcting codes with minimum AWGN.
In [12] L. Hanzo, L-L. Yang, E-L. Kuan and K. Yen presented a flexible broadband
mobile wireless communication system based on FH/MC DS-CDMA and reviewed a variety of
existing as well as a range of forthcoming techniques, which might be required for developing
broadband mobile wireless systems exhibiting a high flexibility and high communications
efficiency. We argued that this broadband FH/MC DS-CDMA system exhibits a high grade of
compatibility with the existing CDMA based systems, since it is capable of incorporating a wide
range of techniques developed for the 2nd- and 3rd-generation mobile systems. At the time of
writing research is well under way towards the SDR-based implementation of a range of existing
systems. It is expected that these efforts will soon encompass a generic scheme not too different
from the FH/MC DS-CDMA scheme advocated here.

In [13] T. Pollet.at.al discussed the following: the common technique of parallel

transmission on many carriers; the performance that can also be achieved on an undistorted
channel; algorithms for gaining that performance; interacting with channel impairments;
modification of the performance through coding; and methods of working section. Also discuss
duplex operation of MCM and the probable use of this on the GSTN.The performance of an
OFDM system is being restricted by AWGN. In literature many methods have been proposed to
decrease the effect of AWGN from OFDM system.
In [14] H. Nikookar and R. Prasad the equalization of the received OFDM signal in
frequency domain which is given gained by the assumption that the channel impulse response is
changing in a linear fashion during a block period and compensate or fulfilled for the ICI terms
that affect the bit-error rate (BER) performance.
In [15] Wei Chen, technique can be used in which Nyquist function is used to mitigate
AWGN. Which has the probability conserves the peak power for the OFDM system. The coding
may be adaptive in case there is unconsumed" guard interval and peak power that exploits the
samples in this region to mitigate additive noise and AWGN effect.
In [16] Chang R. W. theory for cancel the effect of AWGN. In PTS, one data symbol
modulates onto the next sub carrier with predefined weighting coefficients. By doing so, the
PAPR signals generated within a group can be each other and canceled out AWGN in.
In [17] OFDM spectrum the sub carriers consist of a main lobe followed by reducing
amplitude side lobes. There is no interference among the sub carrier as long as orthogonally is
maintained, because at the peak of every sub carrier there exist a spectral null. Frequency offset
leads to loss of orthogonally because the spectral null does not coincide of the individual
carriers peak.
In [18]R.R. Mosier and R.G. Clabaugh for reduction of AWGN,PAPR & ICI power from
OFDM system Coded OFDM was proposed in which the frequency separation between adjacent
sub carriers is increases. In order to realize this goal firstly, two adjacent sub-carriers are paired
as a group; then, jointly code data information in them.After coding, only one of the two sub-
carriers in that pair is transmit over each OFDM symbol interval. As in the coded OFDM system,
only half sub-carriers are used for information transmission.
In [19] propose a 32-point FFT & IFFT design for communication application like
OFDM. The main objective of proposed architecture is to design efficient multiplication of FFT

& IFFT using Vedic multiplication. It has numerous advantages such as: increase the speed,
efficient timing, and better resource utilization parameter. In summary, speeds performance of
our design easily satisfies most application requirements based on OFDM modulated wireless
communication system.
2.2 Problem Formulation
In this thesis, I have generated the sequences to provide the effective code word for
OFDM transmission by using the previous work done by various and used their work to reduce
the AWGN to the minimum possible level at no cost of BER and system performance. Also the
codes provide the good error correcting capability.
It has been presented the simulation results of the scheme through MATLAB coding. The
results show the significant reduction in AWGN with reduced BER and increased system
performance. Although OFDM has some advantages that make it suitable for fading channels, it
presents a high AWGN, which is one of the main drawbacks of OFDM systems.
A simple technique used to reduce the AWGN of OFDM signals is to clip the signal to a
maximum allowed value, at the cost of BER degradation and out-of band radiation without
degrading the BER but at the cost of adding extra information
The goal of this thesis is to select the sequence. The sequences solves the notorious
problem of power control in OFDM systems by maintaining a peak-to-mean envelope power
ratio at minimum possible level while allowing simple encoding and decoding at high code rates
for binary, quaternary or higher-phase signaling together with good error correction.
The whole scheme is programmed with MATLAB R201a to show the Simulation
Results. The graph generated after simulation shows the reduction in AWGN and Bit error rate in
comparison to SNR.
2.3 Simulation Tool
MATLAB (Matrix Laboratory) is a numerical computing environment and fourth-
generation programming language. Developed by Math works, MATLAB allows matrix
manipulations, plotting of functions and data, implementation of algorithms, creation of user
interfaces, and interfacing with programs written in other languages, including C, C++, Java, and
FORTRAN.
MATLAB R2013a version is used for simulation purpose. The communication toolbox in
MATLAB is used for calculating BER under dierent modulation schemes and optimization

toolbox is used for maximize the function which is based on Trust-region algorithm.
Using MATLAB, simulation of OFDM was done with different modulation techniques using
TSM technique. The digital modulation schemes such as QAM were selected to assess the
performance of the designed OFDM system by finding their bit error rate (BER) for different
values of signal to noise ratio (SNR). In this project, I have focused on the problem of AWGN
reduction with standard OFDM system. Different modulation techniques are considered for
PAPR reduction and compared with each other for their performances. I have also briefly
discussed OFDM and its advantages and disadvantages.
2.4 Overview of This OFDM Simulation Project

When the simulation proceeds through the OFDM transmitter and
communicationchannel, it pauses and waits for the user to trigger for proceeding to the receiver.
Thereason for using the last two *mat files is that as soon as the OFDM receiverproceeds, the
program will clear all data/variables stored in MATLAB workspace.
Since MATLAB has a built-in function which performs Inverse FastFourier Transform,
IFFT is opted for the development of this simulation. Six m-filesare written to develop this
MATLAB program of OFDM simulation. One of them isthe main program script file, which is
the only file that needs to be run, while otherm-files will be invoked accordingly. A 512-
grayscale bitmap image is required as thesource input. Another bitmap image file will be
generated at the end of thesimulation as the output. Three MATLAB data storage files
(err_calc.mat,ofdm_parameters.mat, and received.mat) are generated during the
simulation.(err_calc.mat)is to archive the baseband data before the transmission, and be
retrievedat the end of the simulation for the purpose of error calculations.OFDM
parameters.matis to archive the parameters initialized at the beginning of thesimulation and
reserve them for the receiver to use later.

CHAPTER 3
OVERVIEW OF THE OFDM SYSTEM

In this chapter, basic information of OFDM and its detailing of block diagram is shown,
how guardtime and cyclic extension can be used to avoid this inter symbol interference and brief
discussiononAWGN problem in OFDM system.
3.1.Ofdm Introduction
In OFDM system, a high-data rate channel can be divided into number of N number of
low data-rate sub-channels and each and every sub channel can be modulated in different and
varied sub-carrier. Thoselow datarate sub channels have bandwidth less than that the coherence
bandwidth of the channel.
On carrying out this so each and every sub channel have a flat-fading and equalization at
the receiver isminimum amount complexity. By choosing a set of (orthogonal) carrier
frequencies of special kind, high spectralefficiency can be obtained becauseof the spectra of the
SCs overlapping, while on mutual influenceamong the SCs are avoided.
In an OFDM environment the input bit can be multiplexed into number of N symbol,
each and every with symbolperiod of T, and each symbol stream can be used to modulate the
parallel sub carriers. Thesub carriers can be used in this separatedby 1/NTs in frequency domain,
so they are used as an orthogonal over(0, Ts).
A typical OFDM transceiver system is shown in fig 3.1initially, serial to parallel
converter converts the input bits stream into a set groups ofthe log2Mbits, where in which M
may alphabet of size of digital modulation scheme are being used in different sub carrier. Overall
N symbols X (k) can be formed andcreated. Then, the N symbols can be mappedto IFFT. These
IFFT corresponds to the orthogonal sub-carriers in the OFDM symbol.

Fig3.1 OFDM Signal AndTransceiver System
j2mn
1
x(n) = N N1
M=0 Xmxe N 0nN1 (3.1)
Where we say Xm can be baseband symbols on our each sub-carriers. The digital-to-
analog (D/A) converter then leads to an analog time-domain signal which can be transmitted
through the channel.The digital-to-analog (D/A) converter leads to an analog time-domain signal
which can betransmitted through our channel. Before transmitting the OFDM symbol cyclic pre-
fixmustalso be appended at the facing end of symbol. At our receiving end, the cyclic prefix
getremoved then the signal can be converted before its back to a discrete N point sequence y (n),
for eachsub-carrier. This discrete signal can be demodulated by using an N-point fast Fourier
transform(FFT) operation at the receiver. The demodulated symbol stream is given by:
Y(m) = N1
n=0 y(n)e
j2mn/N
+ W(m) 0 m N 1 (3.2)
Where, W (m)shown to the FFT of the sample of w (n), which is itsAWGN channel.
3.2 OFDM Principle

In OFDM system, there are mainly two principles are being used as follows:
1. IFFT and FFT are being used for that respectively, modulating and demodulatingthe data
constellations over orthogonal SCs [1]. These signal processingalgorithms substituting the banks
of I/Q-modulators and demodulators that be required.

Focus on the input of the IFFT, Ndata constellation points are present, where there isNis the
number of FFT points. (iis an index on the SC; kis an index on theOFDM symbol). These
constellations aretaken according to any of different phase shiftkeying (PSK) or by QAM
signaling set (symbol mapping). The number of Noutput samplesof the IFFT, on becoming
TDform the baseband signal carrying the data symbols ona set of Northogonal SCs. In our real
system not all of these N possible
SCs can be used for data.
2. The second important principle is of introduction of a cyclic prefix as a GI, whose
lengthshould reach the maximum excess delay of the multipath propagation channel [1].Owing
to the cyclic prefix, the transmitted signal transform into periodic, and the effect ofthe time-
dispersive multipath channel changes into equivalent to a cyclic convolution,removing the GI at
the receiver. Because of the properties of the cyclic convolution,the effect of the multipath
channel becomes limited to a point wise multiplication of thetransmitted data constellations by
the channel TF or FT of the channel IRthat is, the SCs that remain orthogonal. The bad outcome
of this principle is thedecrement of effective transmits power, during redundant GI are
transmitted. Most of the time GI is picked of a length of one tenth to a quarter of the symbol
period,give rise to an SNR loss which is of 0.5 to 1 db.
The equalization needed for determining the data constellationsbecomes an element wise
multiplication of FFT output by inverse of estimatedchannel TF (channel estimation).
3.3 Guard Time and Cyclic Extension
In order to remove the inter symbol interference completely; guard time can be
introducedfor each OFDM symbol. The guard time should be larger than the expected Delay
spreadof multipath channel. The guard time has no signal. In that case AWGN, PAPR &ICI
would arise. AWGN,PAPR &ICI can be the crosstalk between sub carriers andhence sub carriers
no longer orthogonal. This effect is being shown in Fig. 3 2. Asub carrier one and delayed sub
carrier two are shown. When an OFDM receiver perform tomodulate the first sub carrier, it will
have some interference from the second subcarrier, because in FFT interval, no integer number
of cycles differencebetween sub carrier one and two. Along with this, there will crosstalk from
the first tothe second sub carrier for the similar reasons. For removingAWGN,PAPR &ICI,
OFDM symbol should be cyclicallyextended in the guard time as in Fig. 3.3 this confirms that
the delayed replicas associated with OFDM symbol always have an integer number of cycles in

the FFT interval,as the delay is smaller than the guard time. Fig. 3.4 shows the OFDM
spectrumwhich will fulfill Nyquist condition and criterion for AWGN,PAPR& ISI free pulse
shape. AsOFDM receiver determine and calculates the spectrum values at those points that
corresponds to themaxima of individual sub carriers, so it can demodulate the every single sub
carrier free from anyinterference from other sub carriers. Therefore, instead of ISI, its an ICI
that is being avoidedby having the maximum with respect to one sub carrier spectrum
corresponds to zero crossing associated to all others. Appearance of AWGN is such a big
problem in OFDM related system.
ICI on subcarriersubcarrier#1
Delayed subcarrier#2
AWGN
Guard time FFT integration =1/carrier spacing
OFDM symbol time

Fig 3.2.: Effect of Multipath with Zero Signal inthe Guard Time; The Delayed Sub
Carrier 2 Causes ICI on Sub Carrier 1
Guard time /cyclic prefixFFT integration time =1/carrier spacing
OFDM symbol time

Figure 3.3.: OFDM symbol with cyclic extension

Figure 3.4.: Spectra of individual sub carriers
3.4 Problems Associated In OFDM System

With respect to OFDM link, if transmitter and receiver frequency completely matches
then it is only thesub carriers orthogonal. Any of frequency offset results within the AWGN. A
real oscillatordoes not produce a carrier at absolutely one frequency, but a carrier which isphase
modulated by random phase jitters. As a result the frequency, which is the timederivative
concern with phase, is never constant and hence gives rise to level increase ofAWGNin OFDM
system.

CHAPTER 4
OFDM TRANSMITTER AND RECEIVER

4.1. Proposed OFDM System

This section aims to provide principal and theoretical background to digital
modulation,and different digital band pass modulation techniques. This section will present the
step by step results for the designed fundamental OFDMmodem. As depicted in Figure 4.1, the
simplified OFDM transceiver is divided intothree main sections of Transmitter, Channel and
Receiver, results of which arepresented below. Modulation/De-modulation, IFFT/FFT, CP
insertion/removal are themost important blocks in this simplified OFDM transceiver.
Fig 4.1 OFDM Trans-receiver

In order to simulate this OFDM modem channel bandwidth of 1 MHz with 128subcarrier
is being selected. In addition, symbol duration of 128 s and guard intervalof 20 s have been
chosen.
4.2. Digital Modulation
The process, in which digital symbols get transformed into signals or waveforms thatare
compatible with the channel characteristics, is known as the digital modulation. Two of the most
common types of digital modulation are baseband modulation andband pass modulation. In the

baseband modulation, waveforms mostly take the form ofshaped pulses whereas in the band pass
modulation the shaped pulses modulate asinusoid called a carrier wave, or simply a carrier, out
of which band pass modulationare well known and often used for their important benefits in
signal transmission .There exist other modulation schemes one of which is known as spread
spectrummodulation. In this type of modulation schemes, the systems bandwidths are requiredto
be greatly larger than the bandwidth required by the message (i.e. the minimumbandwidth). Such
class of modulation schemes is responsible for ;
1. Separating the signals if more than one signal uses a single channel
2. Minimizing the effects of interferences
3. Placing a signal in a frequency band where it is required by the design
The digital band pass modulation techniques will be considered in this project.
4.3. Digital Band Pass Modulation Techniques

Band pass modulation is one of the most common types of digital modulation, in
whichthe information signal gets converted to a sinusoidal waveform. Band pass modulationcan
either be analogue or digital modulation, in which the sinusoid duration (T) isdenoted as a digital
symbol.
All sinusoidal signals have three main features of amplitude, phase, and
frequency,making them distinguishable from one another. The process at which any change
inamplitude, phase or frequency of the carrier or combination of the three features, is inline with
the data to be transmitted, is defined as band pass modulation.
s(t) = A(t)cos(t)
A carrier signal can have a general Equation 4.1:
s(t) = A(t)cos(t) (4.1)
The time-varying angle and can be written as Equation 4.2:
(t) = ot +(t)(4.2)
By substituting Equation 4.1 into Equation 4.2 it will produce:
A(t)cos[ot+(t)](4.3)
Where:
A(t) the time-varying amplitude
w0 the radian frequency of the carrier

(t)the phase
It is also worth mentioning that the terms w (frequency in radians per second) and f(hertz) both
represent the frequency, where they are linked by:
w = 2f
The principal ofband pass modulation and demodulation can be classified under
twogroups of coherent and non-coherent modulation/demodulation.
Coherentmodulation/demodulation is a process in which the receiver detects the signal by the
knowledge of the carriers phase. The Frequency Shift Keying (FSK), Amplitude Shift Keying
(ASK), Phase Shift Keying (PSK), Continuous PhaseModulation (CPM) and the hybrid
combinations can be listed as the coherentModulation/demodulationtypes.
On the other hand, the non-coherent modulation/demodulation is a process where
suchphase reference information is not exploited by the receiver. Clearly out of the two,the non-
coherent system is most advantageous as it has lower complexity. TheFrequency Shift Keying
(FSK), Amplitude Shift Keying (ASK), Differential Phase ShiftKeying (DPSK), Continuous
Phase Modulation (CPM) and the hybrid combinations canbe listed as the non-coherent
modulation/demodulation types. Although it waspreviously indicated that phase information is
not utilized in the non-coherent process,but the pseudo-PSK known as the DPSK, is an
important form of non-coherent PSKas it takes the prior symbols phase information in order to
detect the current symbol.
4.4. Phase Shift Keying
Phase shift keying (PSK) is one of the main digital band pass modulation techniques
andwas developed during the early periods of outer-space program. This modulationtechnique is
now extensively used in many systems and applications such as military,commercial
communication systems etc.
The generic mathematical expression for PSK is
2E
cos[ot +(t)] 0< t < T i=1,(4.4)
T
Where:
E: The symbol energy
T: The symbol time duration (0 t T)
The i(t) is the phase term and can be expressed as:

i(t) = 2i/M i=1.M(4.5)

One of the most common examples of PSK modulation technique is the Binary PSK(BPSK)
where M is 2.
4.5. Frequency Shift Keying
Frequency Shift Keying (FSK) is another main type of digital band pass modulationtechnique,
and its general mathematical and analytical expression is as follow :
2E
Si (t) = cos[it +] 0<t < T i=1M
T
Where:
frequency term and has M discrete values
The phase term and it is an arbitrary constant
E The symbol energy
The symbol time duration (0 t T)
In a typical FSK modulation, the change in the binary state of the carrier signal causes achange
in the frequency of the signal. The modulation techniques with such behavior are classified
asContinuous-Phase FSK (CPFSK). It is important to note that in a general M-FSKmodulation
technique, such changes usually take place rather sudden, as the phase is notrequired to be
continuous. M equals 3, whichcorresponds to the number of waveform types. In fact, M value
are selected as a non-zeropower of 2 (i.e. 2, 4, 8, 16 etc).
4.6 Amplitude Shift Keying

Amplitude Shift Keying (ASK) is another main type of digital bandpass
modulationtechnique, and its general mathematical and analytical expression is as follow:
2E
Si (t) = cos[it +] (4.7)
T
i= 1, 2, .M
0tT
Where:
Ei(t)
The amplitude term which will have M discrete values T
The phase term and it is an arbitrary constant

T symbol time duration (0 t T)

It can also be observed that the vector representation of the ASK modulation type is the
same as the one for the PSK modulation, containing S1 and S2 which correspond to the zero
amplitude and a maximum amplitude respectively.
At the beginning of the twentieth century, initial and special type of ASK wasdeveloped.
This special modulation type was usually referred to as Binary ASKsignaling or on-Off Keying
(OOK) and was commonly used in radio telegraphy.Since the ASK is not used as broadly as
before in digital communication systems, andtherefore it will not be used as a modulation type
for the simulation of OFDM system.
4.7 Quadrature Amplitude Modulation (QAM)
In addition to the 3 basic types of modulation techniques, there are other
modulationtechniques that are created by combining these three basic techniques. One of the
mostimportant and popular types of these combinational techniques are QuadratureAmplitude
Modulation (QAM). ASK is combined with PSK to create this hybrid systemknown as QAM,
where both the amplitude and phase are changed at the same time.
QAM modulation techniques are often employed in many radio communications,
datadelivering related or broadcast applicationsa mathematical, analytical and vectorpresentation,
as well as the waveform representation of such modulation technique isincluded.
A method for implementing QAM modulation is through use of rectangularconstellations, which
can be divided into independent Pulse Amplitude Modulated(PAM) components, namely for
both the In-phase and the Quadrature (I-Q) part. Anexample of rectangular constellations for
Quadrature Phase Shift Keying (QPSK),16QAM and 64QAM is illustrated in Figure 4.2 showing
that the constellations are notnormalized. So as to normalize the constellations to an average
power of one, they willneed to be multiplied by a normalization factor, list of which can be
found.
This table also contains the BPSK (2PSK) modulation, which only uses two of the
fourQPSK (4PSK) constellation points.

Figure 4.2QPSK, 16-QAM and 64-QAM constellations
4.8 Use of Basic Types of Digital Modulation Techniques inOFDM

This part will explain and illustrate the use of the above explained basic
digitalmodulation techniques in an OFDM. In doing so, a high-level technical
computinglanguage called MATLAB was used in order to design and implement the
outlinedOFDM communication system. The codes for these algorithms have been written
usingthe guidelines. The codes for design and its implementationemployed in this project are
provided in both the appendices and the enclosed CD.
MATLAB 2013 is very powerful software that is widely used across all
engineeringdisciplines, supporting various graphical input and outputs. Therefore
producingefficient codes and making the simulation results easily illustratable in
differentgraphical formats. MATLAB is also known an interpreted language for
numericalcomputation as it allows one to perform numerical calculations without the need
forcomplicated and time consuming programming. Another important advantage of
usingMATLAB is that it supports several file formats and extensions (for either Image oraudio
files). This will greatly help and validate the judgments with respect to theachieved results.
4.9 Theory ofAWGNModel and Its Influence onOFDMReceiver

This section aims to provide concise principal and theoretical background to OFDM
andits main units of the OFDM receiver, designed in this research.

4.9.1. The Awgn Channel

A noise can be regarded as unnecessary electrical signals that are constantly found
inelectrical systems. When such unwanted electrical signals get superimposed on a cleansignal,
the product will be an unclear signal. This ambiguity makes it very difficult for the receiver to
decide on the correct symbol, and therefore restricts the rate ofinformation transmission. Noise
can have different sources, natural and manmade.
Natural noise is mainly generated from electrical circuits, component noise
andatmospheric disturbances. Man-made noise, on the other hand, is generated fromsources like
spark-plug ignition, radiating electromagnetic signals and switchingtransients. Although
engineers try to reduce and to some extend eliminate thenoise and its effect by shielding, filtering
and correct selection of modulation, but thereremains one type of natural source, known as
Johnson or thermal noise which can neverbe eradicated. Such natural noise types are a result of
thermal movement ofelectrons in electrical components (i.e. resistors and wires etc.). As
described in Chapter 3, most of the communication systems that use channel as a medium for
transmissionface distortion due to presence of noise in the channel.
A very common mathematical channel model, which has also been implemented in
thisresearch project, is the Additive White Gaussian Noise (AWGN) channel. This channelmodel
has been extensively used in determining the most suitable modulation type,modulation order
and comparison between the different encoding schemes.The AWGN is a well-recognized
channel model as it represents a physical reality,provided that the thermal noise at the receiver is
the only source of interference.Nonetheless, due to their simplicity, AWGN are commonly
characterized to symbolizefor a man-made noise or other multiuser interferences.
In AWGN channel, a white Gaussian noise is added to a real or complex input
signal.Depending on the real or complex format of the input signal, the Gaussian noise addedwill
be either real or complex, therefore producing a real or complex output signalrespectively;
inheriting its sample time from the input signal. This channel model usesthe signal processing
random source for generating the noise. In doing so, the randomnumbers are generated using the
Ziggurat method, which is the same method used bythe MATLAB random function.
A Probability Distribution Function (PDF) of a Gaussian distributed random variable
n,with mean value of , and the variance of can be written as Equation 4.8:

1
exp [1/2(n/)2] (4.8)
2
Where:
P(n): PDF of a Gaussian distributed random variable n
2The Variance of n
The normalized Gaussian density function can be obtained when the mean value () equals 0,
and = 1. The normalized PDF is shown in Figure 4.3.
0.399
= 1
.3
.242
probability p(x)
0.2
0.1
2
0.054
-3 -2 -1 0 1 2 3
Fig 4.3 Normalized = 1 Gaussian probability density function of an AWGN channel

The noise n (t) is an additiverandom disturbance of the useful transmitted signal S (t),
illustrated in Equation 4.2.
This noise is categorized as whitenoise, as its Power Spectral Density (PSD) is constant.
The noise has Gaussiandistributed random variable with mean of zero and stationary
characteristic.
The received signal in the time interval of 0 0 t T can be expressed as Equation
4.9Figure 4.2 shows the typical model for the transmitted signal passed through an
AWGNchannel and received signal out of the channel [25].
r(t)= sm(t)+n(t) (4.9)
Where
r (t): Received SignalSm(t): Transmitted Signal

n (t): Sample function of the AWGN process with power density function
Channel
Transmitted signal Sm(t) Received signal r(t)
Noise n(t)
Fig. 4.4Typical model for the transmitted signal passed through an AWGN
Channeland Received Signal Out Of The Channel

The Cumulative Distribution Function (CDF) of a Gaussian or normally distributedrandom
variable can be presented as Equation 4.4 [6]:
1 x
F(x) = 2[1 +erf ( )](4.10)
2.
The CDF of a Gaussian distributed random variable with the mean value = 0 andvariance of 2
=1
4.9.2. Signal-to-Noise Ratio (SNR) Degradation

The SNR value is used in order to measure the performance of the channel at
differentpoints, and to compare the level of a desired signal to the level of background noise.
This value can be calculated as a ratio of average signal power and average noise
power,as it is likely that an instantaneous signal to noise ratios will be considerably different.A
ratio higher than 1:1 indicates that there are more signal than noise. Both signal andnoise power
must be measured at the same and equivalent points in the system, andwithin the same systems
bandwidth. The SNR power equation is defined in Equation (4.11) [7]:
signal power
SNR= noise power (4.11)
Due to a very wide dynamic range of many signals, SNR values are often expressed byusing the
logarithmic decibel scale (i.e. dB). The SNR in dB can therefore be defined as:
signal power
SNRdB =10log10( noise power )= Signal Power dB Noise Power dB(4.12)

It is worth noting that if the signal and the noise are measured across the sameimpedance,
then the SNR can be obtained by calculating the square of the amplituderatio as:
signal power P signal A signal 2
SNR= noise power = P Noise=( A noise ) (4.13)
A = The Root Mean Square (RMS) amplitude

The SNR defined in Equation (4.6) can equivalently be written in the decibel format as:
A signal 2 A signal
SNRdB=10log10( A noise ) =20log ( A noise )(4.14)
The signal can have different types; for instance it can either be as a modulated carrier,baseband
waveform or an information signal. A decrease in the power of the desiredsignal or an increase
in the power of the noise signal can cause degradation to the SNRvalue.
4.10Multicarrier De-Modulation
The task of multicarrier de-modulation at the receiver is based on the
orthogonalitybetween each subcarrier. As it is illustrated in the schematic view in Figure 4.4,
thedemodulation block contains number of _c matched filters, which are responsible
forimplementation of the relation presented in Equation
(m+1)Ts
mTs s(t)k (t mTs).(4.15)
An OFDM system with number of oscillators in the transmitter and number of
matchedfilters in the receiver has a complex implementation for large number of
subcarriers.However, as it was indicated in, if the subcarrier numbers (Nc) is a power of two,then
a cheaper approach may be used. The number of oscillators in the transmitter andnumber of
matched filters in the receiver can therefore be replaced by an IDFT and DFToperation. In
addition to being a cheaper approach, such implementation does not sufferfrom the inaccuracies
that an analogue oscillator bank faces.
Yo,m
20 . .
Ts
Y1 m Y
. .
21

Fig. 4.5OFDM demodulation [13]

4.11. Serial ToParallel Conversion

At this stage, the input serial data stream is formatted into suitable size required
fortransmission, for example one bit for BPSK (21 = 2) or two bits for QPSK (22 = 4),which are
shifted into a parallel format. Following the serial to parallel conversion ofdata, they are then
transmitted in parallel format by assigning each suitably selected sizeto one carrier in the
transmission.
4.12. FFT Implementation

As previously stated, the IFFT and FFT blocks are accounted as the most importantblocks
within the OFDM communication system, as OFDM Orthogonality is providedthrough the use of
these blocks.The IFFT block in the transmitter is responsible for transformation of both
amplitudeand phase of each component of the spectrum into the time domain. As it was
explainedin the transmitter section, Chapter 3, the number of complex data is converted to
thesame number of data in the time domain. In the receiver, FFT block is similarlyresponsible
for the reverse operation; the conversion of the data points from the timedomain back to the
frequency domain. Following the FFT operation, the demodulationof each received phase is
performed. This is carried out through assessing andconversion of the phase angle of each carrier
back to the data word. The achieved datawords are later combined to create the original data with
the same word size.
4.13. Cyclic Prefix Removal

Introduction of cyclic prefix maintains the orthogonality between each of the
subcarriersand also preserves the independence of the following OFDM symbols. As
previouslyexplained in Chapters 2 and 3, cyclic prefix is a percentage copy of the last part of
theOFDM signal, which is affixed to the transmitted signal. Addition of this percentagecopy
results in circular convolution between the channel impulse response and thetransmitted signal,
making its frequency domain a multiplication of frequency responsefor both channel and
transmitted signal. This makes the received signal in itsfrequency domain to be a scaled form of
the transmitted signal in its frequency domain.Therefore, formation of any distortion due to the
severe channel conditions is eliminated.

The different blocks within the receiver perform a reverse operation to the transmitter.The cyclic
prefix removal is carried out at the receiver subsystem, allowing a cyclicprefixfree signal to be
passed through the different blocks of this subsystem.

CHAPTER 5
SIMULATION AND RESULT

5.1. Simulation of Transmitter

The simulation of OFDM transmitter is discussed in this section, containing the step bystep
results and discussions. The architecture of this particular OFDM transmitter isdepicted in Figure
5.1.
Serial to BPSK IFFT bins IFFT Cyclic Cascade

parallel QPSK extention
DAC
allocation frame add
Source Frames conversio QAM addition frame
Data from divider n Modulation Gaurds
input bmp header
file
RF
andtrailer
front-
end
channel
Communication
Fig 5.1 OFDM transmitter
As depicted and highlighted in Figure 5.1, the initial stage prior to the actual
OFDMtransmission is to transmit the generated message, where this message could be
eitherrandomly generated binary values, audio sound, or digitally processed image.Simulation of
this part uses the uniformly distributed pseudorandom numbers, by usingthe rand(m,n)
function which produces a 1 by 2500 pseudorandom values, where2500 represent the number of
bits. In order to produce random binary values, thepreviously produced values are required to be
rounded to their nearest integer value.This is achieved by the use of round function, which
produces 1 by 2500 bits (zerosand ones). Figure 5.1 depicts the random binary generated
message.
It is important to note that the input message illustrated in Figure 5.1 and the followingfigures
presenting an output for each block, is only part of the full message, andtherefore only gives an
indication of 12 bits out of 2500 bits.
The core of the OFDM transmitter is the modulator, which modulates theinput data
stream frame by frame. Data is divided into frames based on the variable symbol per frame,
which refers to the number of symbols per frame per carrier. It isdefined by: symb_per_frame =
ceil(2^13/carrier_count). This limits the totalnumber of symbols per frame (symb_per_frame *

carrier_count) within theinterval of [2^13, 2*(2^13-1)], or [8192, 16382]. However, the number
of carrierstypically would not be much greater than 1000 in this simulation, thus the total number
of symbols per frame would typically be under 10,000. This is anexperimentally reasonable
number of symbols that one frame should keep under forthis MATLAB program to run
efficiently; thereby symbol perframe is defined by theequation shown above. If the total number
of symbols in a data stream to betransmitted is less than the total number of symbols per frame,
the data would not bedivided into frames and would be modulated all at once.
Following the simulation results for the main types of modulation scheme, the
principaland theoretical background to the main blocks of an OFDM transmitter, including
basicprinciple of OFDM, implementation of IFFT, and cyclic prefix were presented.
The stated basic digital modulation techniques in an OFDM were next illustrated. Indoing
so, a high-level technical computing language called MATLAB was used inorder to design and
implement the outlined OFDM communication system. Morespecifically, a step by step
simulation response for each block of the transmissionsection of the designed fundamental
OFDM modem, consisting of the most importantblocks such as Modulation, IFFT and CP
insertion were presented. Summary of thissequential step by step simulation of the transmission
section are as follows:
1) The random binary message were generated
2) These binary message were converted from serial format into a parallel
3) The parallel binary message were modulated using the BPSK
4) The IFFT were applied in order to find the corresponding time waveform
5) The cyclic prefix was added by copying a percentage of the OFDM symbol and Pre pending it
to the transmitted symbol.
Figure 5.1 shows a block diagram of a generic OFDM system. ADC, DAC, andRF front-ends
(Amplification, RF upconversion/downconversion, etc.) are notsimulated in this project. This
MATLAB simulation program consists of six files.
OFDM_SIM.m shall be run while other m-files will be invoked accordingly.Source data
for this simulation is taken from an 8-bit grayscale (256 graylevels) bitmap image file (*.bmp)
based on the users choice. The image data willthen be converted to the symbol size
(bits/symbol) determined by the choice of MPSKfrom four variations provided by this
simulation. The converted data will thenbe separated into multiple frames by the OFDM

transmitter. The OFDM modulatormodulates the data frame by frame. Before the exit of the
transmitter, the modulatedframes of time signal are cascaded together along with frame guards
inserted inbetween as well as a pair of identical headers added to the beginning and end of
thedata stream. The communication channel is modeled by adding Gaussian white noiseand
amplitude clipping effect.
The receiver detects the start and end of each frame in the received signal byan envelope
detector. Each detected frame of time signal is then demodulated intouseful data. The modulated
data is then converted back to 8-bit word size data usedfor generating an output image file of the
simulation.
5.2 Simulation of Receiver
The simulation of OFDM receiver is discussed in this section, containing the step bystep
results and discussions. The architecture of this particular OFDM receiver isdepicted in Figure
channel
Communication
5.2.
Received
data to
Extract
generate Cyclic
Cascade Parallel BPSK QPSK carrier Frame
output bmp FFT extension Serial
frame to serial QAM from detection
file removal to RF
Demodulatio FFT bins
parell ADC front
n
el end
Fig 5.2 Receiver

The OFDM receiver basically does the reverse operation to its transmitter, whereinitially
the guard interval is detected and then removed. Later, the FFT of each symbolis then taken to
find the original transmitted spectrum. The phase angle of eachtransmission carrier is then
evaluated and converted back to the original transmitted databy demodulating the received
phase. The suitable data size received are then combinedand re-formatted back to the original
serial data stream.
As depicted in Figure 5.1, the initial stage for OFDM reception is to receive the
outputsignal from the channel, where the signal is corrupted with applied noise within thechannel

that needs to be removed. The received input signal to the OFDM receiver, andremoval of noise
is shown in Figure 5.2.
A trunk of received signal in a selective length is processed by the framedetector
(ofdm_frame_detect.m) in order to determine the start of the signal frame.For only the first
frame, this selected portion is relatively larger for taking the headerinto account. The selected
portion of received signal is sampled to a shorter discretesignal with a sampling rate defined by
the system. A moving sum is taken over thissampled signal. The index of the minimum of the
sampled signal is approximatelythe start of the frame guard while one symbol period further
from this index is theapproximate location for the start of the useful signal frame. The frame
detector willthen collect a moving sum of the input signal from about 10% of one symbol
periodearlier than the approximate start of the frame guard to about one third of s symbolperiod
further than the approximate start of the useful signal frame. The first portion,with a length of
one less than a symbol period of this moving sum is discarded. Thefirst minimum of this moving
sum is the detected start of the useful signal frame.
System Configurations and Parameters

Table 5.1 Parameter Table
Parameters Values
Source image size 1200*1500
IFFT size 2048
Number of carriers 1024
Modulation methods BPSK QPSK QAM
Peak power clipping 12 db
Signal to noise ratio 9db

At the beginning of this simulation MATLAB program, a script fileofdm_parametersmis

invoked, which initializes all required OFDM parameters andprogram variables to start the
simulation. Some variables are entered by the user.
The rest are either fixed or derived from the user-input and fixed variables. The
userinputvariables include:
1) Input file an 8-bit grayscale (256 gray levels) bitmap file (*.bmp);
2) IFFT size an integer of a power of two;
3) Number of carriers not greater than [(IFFT size)/2 2];
4) Digital modulation method BPSK, QPSK, QAM.
5) Signal peak power clipping in dB;
6) Signal-to-Noise Ratio in dB.
The number of carriers needs to be no more than [(IFFT size)/2 2], because thereare as
many conjugate carriers as the carriers, and one IFFT bin is reserved for DCsignal while another
IFFT binis for thesymmetrical pointat the Nyquistfrequency toseparate carriersand conjugate
carriers. All user-inputs are checked for validity and the program willrequest the user to correct
any incorrect fields with brief guidelines provided. Anexample is shown in Table 1. This script
also determines how the carriers andconjugate carriers areallocated into theIFFT bins, based
onthe IFFT size andnumber of carriersdefined by the userconjugate carriers spreading out on
512IFFT bins.
Two properties of a typical communication channel are modeled. A variableclipping in
this MATLAB program is set by the user. Peak power clipping isbasically setting any data points
with values over clipping below peak power toclipping below peak power. The peak-to- RMS
ratios of the transmitted signalbefore and after the channel are shown for a comparison regarding
this peak powerclipping effect.
Channel noise is modeled by adding a white Gaussian noise (AWGN) defined
By:
(variance of the modulated signal)
AWGN= linearSNR
It has a mean of zero and a standard deviation equaling the square root of the quotientof
the variance of the signal over the linear Signal-to-Noise Ratio, the dB value ofwhich is set by
the user as well.
5.3 Proposed Simulation Prosess inMATLAB

Seven graphs are plotted during this OFDM simulation:
1. Magnitudes of OFDM carrier data on IFFT bins;since all magnitudes are ONE, what this plot
really shows is how thecarriers are spread out in the IFFT bins.
2. Phases translated from the OFDM data;In this graph, its easy to see that the original data has
a number ofpossible levels equal to 2 raised to the power of symbol size.
3. Modulated time signal for one symbol period on one carrier +4 Modulated time signal for one
symbol period on multiple (limiting to six) carriers.
5. Magnitudes of the received OFDM spectrumthis is to be compared to the first graph.
6. Phases of the received OFDM spectrum this is to be compared to the second graph.
7. Polar plot of the received phases A successful OFDM transmission and reception should have
this plotshow the grouping of the received phases clearly into 2^symbol-sizeconstellations.
The first four plots are derived from OFDM modulation while the last three are
fromdemodulation. None of these plots include a complete OFDM data packet. The firstthree
plots represent only the first symbol period in the first frame of data, whereasthe fourth plot
represents up to the first six symbol periods in the first frame. Sincethe first and last portion of
the received/modulated data have higher probability ofgetting errors due to imprecision in
synchronization, a sample of symbol period usedby the fifth, sixth, and seventh plots is from the
approximate middle of a frame, whichis also approximately the middle one among all data
frames. However, its stillpossible that the sample taken for the demodulation plots is still
erroneous on certaintrials of this MATLAB simulation. It is important to note that even if the
fifth, sixth,and seventh plots dont show reasonable information, the overall OFDM
transmissionand reception would still likely be valid since these plots only represent one
symbolperiod among many.
5.4 Picture Quality versusSignal toNoise Ratio
In addition to the transmission of bits (zeroes and ones) and following the
comparativeperformance studies on the OFDM modem by means of analyzingthose using
differenttechniques and scenarios which has so far been the subject of this chapter; the next
stageis the transmission of an image which will be considered in this section.

Furthermore, this section will illustrate the work undertaken in order to improve
thedegradation performance of OFDM system in the presence of high noise in the channel.
Theimage sent in OFDM system in noise channel. As it is illustrated in this table, the first
image in the lefthand side column is the original input image, and the image next to it is the gray
scaleimage of the original input image. The gray scale image is the image to which thedifferent
techniques such as the convolution encoding/decoding, modulation/demodulation in the
transmitter and receiver will be applied.
As it is illustrated in Table 5.1, like many other digital communication systems,
theperformance of this OFDM system is only acceptable, up to some critical channel noiselevel.
In other words, if the noise level is raised above that critical level, theperformance of the system
fails very quickly. Such matters may highly affect theperformance of the wireless
telecommunications, where the drops in the signal may leadto diminution in reliability of the
communication.The advantage of the currently designed OFDM system is that, when the channel
isunder a high noise condition, the system outputs a worse image quality rather thancompletely
loosing the transmitted image. This is depicted in the pictures of Table 5with SNR value of 3 or
6 dB. This image transmission through the noisy channel hasbeen simulated by using the same
MATLAB code as for the bits transmission, wherethe signal to noise ratio (SNR) of the channel
is varied from 1dB to 20 dB, with theimage quality measured at 3 dB increments.In this OFDM
system, the FEC technique was also used with the 64 - QAM modulationtechnique which
illustrate that:
Using 64 QAM carries higher data rates, which is essential when dealing withimage
transmission.Modulation techniques such as 16QAM perform better than higher ordermodulation
such as (64 QAM) under high noise channel condition with SNR < 12dB.

FLOW CHART OF MATLAB Start

SIMULATIONON OFDM SYSTEM
Input
image
Different
modulation
method
IFFT
Cascade
frame with GI
If ref>threshold
Yes voltage equal to
modulation after IFFT
No
No
Select image and

compare with different
methods

Fig.5.3 OFDM Carriers on designated IFFT bins

Fig 5.4 Phases of the OFDM modulated Data

Fig. 5.5 OFDM Time Signal(one symbol period in one carrier)

Fig. 5.6Samples of OFDM Time Signals over one symbol period

Fig5.7Magnitude of Received OFDM Spectrum

Fig 5.8 Phase of Receive OFDM Spectrum

Fig. 5.9Received Phases

Fig. 5.10 Image to be TX

Fig. 5.11 16-BPSK Received Image
Data loss in this communication = 0.166667% (4096 out of 2457600)

Total number of errors = 1402 (out of 2453504)
Bit Error Rate (BER) = 0.057143%
Average Phase Error = 13.731679 (degree)
Percent error of pixels of the received image = 0.365560%

Fig. 5.12 64-QPSK Received Image

Average Phase Error = 15.963237 (degree)

Fig. 5.13 QAM Received Image

Average Phase Error = 10.67854(degree)

CHAPTER 6
CONCLUSION AND FUTURE WORK

6.1. Conclusion
The decisive and proactive aim of this research project was to design and implement
anOFDM system for communication, by using simulation of the operation ofvirtual transmitter
and receiver. Importantly the only aim of this research wasto design an OFDM modem for a
communication to proposeand examine a outstanding approach in comparing the various
modulation order, differentmodulation type, implementation of FEC scheme and also application
of different noisetypes and using all for the two modeled channels (AWGN channel) in an
endeavor to understand and know the most suitabletechnique for the transmission of message or
image over communication system.
For simulating an OFDM that associated signal could be corruptedof noise the closed
model in this field was adoption of Power linecoloredbackground noise, but most of the
simulations in this project are the outcome ofAWGN only.This principle aim was achieved by
setting and completing number of objectives. Thefirst was to design a basic OFDM modem by
using BPSK and the second was to designa more comprehensive and decisive OFDM modem for
the purpose of simulation. This was followed and taken by investigating the operation of the
communication link with respect tothe OFDM byutilizing Cyclic Prefix, with the aim of
counteracting the effects of delay such as ISI andICI. The performance related to OFDM modem
was examined by taking the AWGNchannel. An alternates comparative performance studies
were taken by plotting theBER graphs for many simulations for different modulation types,
differentmodulation orders and also different types of channel coding as considered. In addition
to thisadditional analysis was done in order to examine the overall performance related toOFDM
modem. It had been initially achieved by using AWGN and later withbackground noise. At last
the complete functioning OFDM modem wasexamined through the transmission of an image.For
this purpose, a high-level technical computing language called MATLAB wasintroduce for

designing and implementing the outlined OFDM communication system. Table contains the
techniques and methods used in the current research, most important resultswere obtained, one of
the best selected techniques and at last some advantages and disadvantages ofthese techniques.
Outcome related to the comparison between different modulation orders to indicatethat
BPSK is the simplest modulation order to implement, because it offers lower data rates.
This simple modulation order gives out a very good performance in the
communicationsystem and it is more robust and resilient in presence of noise and interferences.
AfterBPSK, 4-QAM in comparison to higher order QAM modulation, shown an excellent
performance. However it produced a very similar results to 4-PSK, but a betterperformance in
the system was observed, due to its lower data rates and its robustness inpresence of noise.
Research on the performance of different modulation types was accomplished bystudying
and examining 4-PSK, 4-FSK and 4-QAM, out of which PSK and QAMshown a better
performance. The 4-PSK and 4-QAM modulation types aresimilarly simple to use and
implement. These modulation types have shown a high performance incommunication system as
they offer a lower data rates and are more robust in presenceof noise. It is also important to note
that when faster data rates are required, 4-QAM becomes abetter option. In fact we can say this
method of comparison between the different modulation typesproved to be a useful method of
testing these modulations.
The results of comparison between different encoding techniques shown
thatconvolutionchannel soft encoding and block channel soft encoding present excellent
performance in comparison to the other channel encoding techniques. Moreover, thiscomparison
has also suggested that when the two encoding techniques areapplied to BPSK modulation; it
provides aacceptable improvement in the OFDMsystem.
In conclusion to this the simulation results presented in this project suggest that BPSK
and 4-QAM, are the most recommended modulation techniques (in both type and order) for
theirconsiderable performance. CCE-Soft and BCE-Soft are the best encodingtechniques (in
FEC type) for the best performance in case of error detection and correction.
In real sense, applying these techniques to the two modeled channels has proved verysuccessful
and will be accounted as a best approach for the transmission of message orimage within a power
line based communication system.
6.2. Future Work

The results oriented summary presented above and contributions made by this research
haveshown that in presence of AWGN and Power linecolored background noisecause a
significant improvement in the performance and after that the transmissionof the system, various
framing techniques such as the use of modulation anderror correction techniques, can minimize
the effect of the noise and interference.
Adding to that the existence of certain assumptions and limitations in this study limited
theout- come and contributions of this work for providing a more accurate comparisonbetween
the techniques, and rightestimation of the errors in actual systems. It is important to notice the
highlight and recommendsome of the areas where if assigned timeand facilities; more research,
investigation and improvements would be excellent.Following is a list of scope for further
research:
With an ideal channel and perfect timing synchronization between thetransmitter and
receiver is not the case related to the real communication systems,but the imperfect conditions
present over the communication system can becorrected by using standard digital signal
processing (DSP) techniques in thetransmitter and receiver ends. This is very important that the
use of these digitalsignal processing techniques oriented to application dependent, if change in
specific parameterof the DSP techniques gives out a different performance. The idea of
synchronization for OFDM communication systems presented in [9,10] can be taken forward
andapplied to the system presented in the current research.It can be made in measuring the noise
in a real life scenario,and examine how best can the signal be taken out as a result of these noise.
One more point which can be introduced to the current research is to implement thediscussed
different modulation schemes (BPSK, QAM etc.) to anotherfading channels that are not shown
in this study, such as Rayleigh, Rician,Weibull and Beckmann channel.
This research emphasized on applying the forward error correctionconvolution channel
encoding and block channel encoding to boast up thesystem performance. but such encoding
systems are the most neededencoding system for real implementation, although the additional
performanceanalysis can be concluded by taking different types of encoding techniques.
Thiscould be executed by using Hamming, Golay and ReedMuller encodingand decoding
techniques, and comparing their performance by used encoding techniques.
The OFDM modem designed in the current research can be used for extensivestudy in
this field at higher level research, with the use of programminglanguages such as C or

usingplatforms such as MATLAB, which will help implementcustomer-based configurable

integrated circuits such as Field-programmableGate Array (FPGA), become feasible and
reconfigurable. In the year 2003, a similarresearch was completed by primarily suppressing the
complexity of NeuralNetworks (NN) for peak to average power ratio (PAPR) reduction related
to OFDMsignal and using it on a FPGA, idea of which could beexplored subsequently
The channel estimation is another research field that would be decisiveinvestigation, or
standardizing the already proposed algorithms, this is worth noticed that the implementation of
OFDM. different concerned research were taken up in theuse of both neural network and the
Cyclic Prefix (CP) based block RecursiveLeast Squares (RLS) algorithm, way to investigate the
channelestimation in an OFDM and SISO-OFDM systems . But thereare number of algorithm
proposed, they remain a great potential for thistechnique to be explored and improved. some
recent investigation have been completed on the channelequalization by using feedback
equalization, blind adaptive equalization andneural network , the research area relating to the
equalization using thecyclic prefix concept could be another point of interest is to extend.Some
researchers have concentrated on change the OFDM weakness another additional analysis is
initially considering completed research and attempting to improve the system.Another point of
interest is to attempt analyzing the performance of thecurrently designed OFDM system in a
multi-user environment and also toestablish a protocol with the ability to conduct acknowledge
and retransmissions.
In the current research, different modulation order and modulation type such as2PSK,
4PSK, 8PSK, 4FSK, 4QAM, 8QAM and 16-QAM were investigated andlater compared. For
future research, up tothe type of data beingtransmitted, the modulation order and modulation
type can dynamically beselected. This would help improving the overall performance, as the
mostsuitable modulation technique for that particular data has been selected.Further research
could also be done on the coding and modulation for communication.
Applying the Artificial Neural Network (ANN) to an OFDM system with theaim of
improving the performance of the system, have been observed in numberof researches and
studiesThis method can therefore be furtherimplemented and applied to different components
within the OFDMcommunication system to improve its performance.

REFERENCES
[1] Patrick J. Langfeld, Klaus Dostert. (2012). OFDM system synchronisation forpowerline
communications. International Symposium on Power LineCommunications (ISPLC). pp. 15
22.
[2] H. Meng, Y. L. Guan and S. Chen. (April 2010). Modeling and Analysis ofNoise Effects on
Broadband Power-Line Communications. IEEE TransactionsOn Power Delivery. Volume 20,
Number 2, pp. 630 637
[3] G. Avril, M. Tlich, F. Moulin, A. Zeddam, F. Nouvel. (2010). Time/FrequencyAnalysis of
Impulsive Noise on Powerline Channels. International Symposiumon Power Line
Communications (ISPLC). pp.143 150
[4] E.Yavuz, F. Kural, N. Coban, B. Ercan and M. Safak. (2012). Modelling ofpowerlines for
digital communication. International Symposium on Power LineCommunications (ISPLC). pp.
161 168
[5] Andrew J. Viterbi and Jim K. Omura (2011). Principles Of DigitalCommunication And
Coding. : McGraw-Hill, Inc. ISBN 0-07-0675, pp. 16-3
[6] C. E. Shannon. (2008). A Mathematical Theory of Communication. The BellSystem
Technical Journal. Volume 27, pp. 379423
[7] F. R. P. Cavalcanti and S. Andersson. (2009). Optimizing WirelessCommunication Systems.
: Springer Science. Page 353
[8] A.R.S. Bahai, B. R. Saltzberg and M. Ergen (2004). Multi-carrier digitalcommunications:
theory and application of OFDM. 2nd edition: SpringerScience. pp. 55 - 68
[9] J. Isabona, and M. E. Ekpenyong. (2010). Data Transmission Of OFDMWireless Networks:
An Optimization Perspective. World Journal of AppliedScience and Technology. Volume 2.
No.1, pp. 87-97.
[10] M. Noh, Y. Lee and H. Park, (October 2011). Low complexity LMMSE channelestimation
for OFDM, IEE Proceeding Communication. Volume 153, No. 5,pp. 645 - 650

[11] Anbal Luis Intini, (2009), Orthogonal Frequency Division Multiplexing forWireless
Networks Standard IEEE 802.11a, MSc Thesis, University OfCalifornia
[12] Dan Raphaeli and Evgeni Bassin. (2011). A Comparison between OFDM,Single Carrier,
and Spread Spectrum for high Data Rate PLC. ISPLC pp. 162 168.
[13] L. Hanzo, L-L. Yang, E-L. Kuan and K. Yen. (January 2003). Single- and Multi-Carrier
DS-CDMA: Multi-User Detection, Space-Time Spreading, Synchronization, Networking and
Standards. University of Southampton andAnalogue Devices Ltd., Cambridge, UK: Wiley-
Blackwell. Chapter 1
[14] T. Pollet, M. van Bladel, and M. Moeneclaey. (April 2007). BER sensitivity ofOFDM
systems to carrier frequency offset and wiener phase noise. IEEETransactions on
Communications,. Volume 43, pp. 191193
[15] H. Nikookar and R. Prasad. (October 2010). On the sensitivity of multicarriertransmission
over multipath channels to phase noise and frequency offset. InProceedings of IEEE
International Symposium on Personal, Indoor, and MobileRadio Communications (PIMRC96).
pp. 6872.
[16] Wei Chen. (October 2005). Time- Frequency- Selective Channel Estimation of OFDM
Systems, PhD thesis, Drexel University, October 2005, pp. 6-19
[17] Richard van Nee, Ramjee Prasad. (2000). OFDM for wireless multimedia communications,
Artech House Inc. pp. 33 55
[18] Chang R. W. (Dec. 1966). Synthesis of Band Limited Orthogonal Signals for Multichannel
Data Transmission, Bell System Technical Journal, Volume. 45, pp. 1775-1796
[19] B.R. Saltzberg. (Dec. 1967). Performance of an efficient parallel data transmission system,
IEEE Transactions on Communication, Volume 15, pp. 805 813
[20] R.R. Mosier and R.G. Clabaugh, (Jan. 1958.) A Bandwidth Efficient Binary Transmission
System, IEEE Trans. Volume. 76, pp. 723 - 728
[21] Mr Abhijit Palekar (Dec 2013) OFDM system Using FFT and IFFT .IJARCSSE Volume
3,Issue 12,December2013

SISO Thesis

Încărcat de

Informații document

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

SISO Thesis

Încărcat de

Drepturi de autor:

Formate disponibile

Performance analysis of orthogonal frequency division multiplexing (OFDM) using different

modulation techniques for image encryption

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 1

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 2

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 3

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 4

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 5

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 6

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 7

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 8

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 9

1.5.Modern Day Wireless Systems

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 10

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 11

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 12

power, it can only be mitigate.

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 13

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 14

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 15

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 16

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 17

In [13] T. Pollet.at.al discussed the following: the common technique of parallel

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 18

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 19

2.4 Overview of This OFDM Simulation Project

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 20

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 21

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 22

Fig3.1 OFDM Signal AndTransceiver System

3.2 OFDM Principle

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 23

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 24

Guard time FFT integration =1/carrier spacing

OFDM symbol time

Guard time /cyclic prefixFFT integration time =1/carrier spacing

OFDM symbol time

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 25

Figure 3.4.: Spectra of individual sub carriers

3.4 Problems Associated In OFDM System

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 26

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 27

4.1. Proposed OFDM System

Fig 4.1 OFDM Trans-receiver

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 28

4.3. Digital Band Pass Modulation Techniques

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 29

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 30

i(t) = 2i/M i=1.M(4.5)

4.6 Amplitude Shift Keying

The phase term and it is an arbitrary constant

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 31

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 32

Figure 4.2QPSK, 16-QAM and 64-QAM constellations

4.8 Use of Basic Types of Digital Modulation Techniques inOFDM

4.9 Theory ofAWGNModel and Its Influence onOFDMReceiver

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 33

4.9.1. The Awgn Channel

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 34

Fig 4.3 Normalized = 1 Gaussian probability density function of an AWGN channel

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 35

Channeland Received Signal Out Of The Channel

4.9.2. Signal-to-Noise Ratio (SNR) Degradation

DIVYA KODWANI, Department of Electrical Engg JEC Jabalpur (M.P.) 36

A = The Root Mean Square (RMS) amplitude