Sie sind auf Seite 1von 60

ABSTRACT

Orthogonal frequency-division multiplexing (OFDM) is a method of encoding


digital data on multiple carrier frequencies. OFDM has developed into a popular
scheme for wideband digital communication, used in applications such as digital
television and audio broadcasting, DSL Internet access, wireless
networks, powerline networks, and 4G mobile communications. However, peak-
to-average ratio (PAPR) reduction and inter-carrier interference (ICI) cancellation
are two major challenges in implementing an OFDM system. High PAPR results
from large envelope fluctuations in OFDM signal which requires a highly linear
power amplifier (PA).

The PAPR of the OFDM signal can also be reduced by using clipper filltering
technique.

1|Page
INTRODUCTION

The modern day phenomenon of increased thirst for more information and the
explosive growth of new multimedia wireless applications have resulted in an
increased demand for technologies that support very high speed transmission
rates, mobility and efciently utilize the available spectrum and network
resources. OFDM is one of the best solutions to achieve this goal and it offers a
promising choice for future high speed data rate systems.OFDM has been
standardized as part of the IEEE 802.11a and IEEE 802.11g for high bit rate data
transmission over wireless LANs.
OFDM is an abbreviation for Orthogonal Frequency Division Multiplexing formed
by the first letter of each word (OFDM). OFDM was rst presented in the late
1950s and characterized in the mid 1960sOFDM is a modulation scheme in
which multiple data bits are modulated simultaneously by multiple carriers. This
procedure partitions the transmission frequency band into multiple narrower
subbands, where each data symbols spectrum occupies one of these subbands.
As compared to the conventional frequency division multiplexing (FDM), where
such subbands are non-overlapping, OFDM increases spectral efciency by
utilizing subbands that overlap. To avoid interference among subbands, the
subbands are made orthogonal to each other, which means that subbands are
mutually independent. By breaking the wide transmission band into narrower,
multiple subbands, OFDM schemes effectively combat the effect of frequency-
selective fading usually encountered in wireless channels. Frequency-selective
fading is a consequence of the phenomenon called multipath propagation, where
multiple copies of the transmitted signal traveling along different paths combine
at the receiver. To overcome the frequency-selective fading, each subband should
be narrow enough such that its bandwidth B satises

2|Page
where av is the average delay spread dened as the average value of the
exponentially distributed random variable used to model the incremental delays
of the multiple received rays of the transmitted signal.

OFDM converts the frequency-selective fading channel into multiple at-fading


subchannels, thereby allows the use of simple frequency-domain equalizers to
overcome the problem. However, OFDM introduces inter-symbol interference
(ISI) and inter-carrier interference (ICI). ISI is the effect adjacent OFDM symbols
exert on each other due to delay spread and ICI is the effect subcarriers exert on
each other. Both of these problems can be reduced signicantly by introducing a
guard interval between OFDM symbols. This interval is a cyclic extension of the
signal itself concatenated at the beginning of the OFDM symbol, called the cyclic
prex (CP). Detailed discussion of the problems of ISI and ICI and the mitigation
techniques used to overcome them are beyond the scope of this survey and will
not be discussed further.

3|Page
4|Page
Theory

Modulation - a mapping of the information on changes in the carrier


phase, frequency or amplitude or combination.

Multiplexing - method of sharing a bandwidth with other independent


data channels.

OFDM is a combination of modulation and multiplexing. Multiplexing


generally refers to independent signals, those produced by different sources.
So it is a question of how to share the spectrum with these users. In OFDM
the question of multiplexing is applied to independent signals but these
independent signals are a sub-set of the one main signal. In OFDM the
signal itself is first split into independent channels, modulated by data and
then re-multiplexed to create the OFDM carrier.

OFDM is a special case of Frequency Division Multiplex (FDM) . As an


analogy, a FDM channel is like water flow out of a faucet, in contrast the
OFDM signal is like a shower. In a faucet all water comes in one big stream
and cannot be sub-divided. OFDM shower is made up of a lot of little
streams.

a b
Fig. 1 (a) A Regular-FDM single carrier A whole bunch of water
coming all in one stream. (b) Orthogonal-FDM Same amount of
water coming from a lot of small streams.

Think about what the advantage might be of one over the other? One
obvious one is that if I put my thumb over the faucet hole, I can stop the

5|Page
water flow but I cannot do the same for the shower. So although both do
the same thing, they respond differently to interference.

Fig. 2 All cargo on one truck vs. splitting the shipment into more than one.
Another way to see this intuitively is to use the analogy of making a
shipment via a truck. We have two options, one hire a big truck or a
bunch of smaller ones. Both methods carry the exact same amount of
data. But in case of an accident, only 1/4 of data on the OFDM trucking
will suffer.
These four smaller trucks when seen as signals are called the sub-carriers in
an OFDM system and they must be orthogonal for this idea to work. The
independent sub-channels can be multiplexed by frequency division
multiplexing (FDM), called multi- carrier transmission or it can be based on
a code division multiplex (CDM), in this case it is called multi-code
transmission.
OFDM is a special case of FDM (just as it says in its name, OFDM)

OFDM is a special case of FDM ( Frequency Division Multiplexing). In FDM, the


given bandwidth is subdivided among a set of carriers. There is no relationship
between the carrier frequencies in FDM. For example, consider that the given
bandwidth has to be divided among 5 carriers (say a,b,c,d,e). There is no
relationship between the subcarriers ; a,b,c,d and e can anything within the given
bandwidth.

If the carriers are harmonics, say (b=2a,c=3a,d=4a,d=5a , integral multiple of


fundamental component a ) then they become orthogonal. This is a special case
of FDM, which is called OFDM (as implied by the word orthogonal in OFDM).

6|Page
OFDM Spectrum

The importance of being orthogonal


Since the carriers are all sine/cosine wave, we know that area under one
period of a sine or a cosine wave is zero. This is easily shown.

Fig. 4 - The area under a sine and a cosine wave over one period is always
zero.

7|Page
Let's take a sine wave of frequency m and multiply it by a sinusoid (sine or a
cosine) of a frequency n, where both m and n are integers. The integral or the area
under this product is given by
f ( t ) sin mwt sin nwt

Figure 5 - The area under a sine wave multiplied by its own


harmonic is always zero.
By the simple trigonometric relationship, this is equal to a sum of two
sinusoids of frequencies (n-m) and (n+m)

1 1
2 cos(m n) 2 cos(m n)

These two components are each a sinusoid, so the integral is equal to zero
over one period.

02 12 cos(m n)t 02 12 cos(m n)t

0 0

We conclude that when we multiply a sinusoid of frequency n by a sinusoid


of frequency m/n, the area under the product is zero. In general for all
integers n and m,sin mx,cos mx,sin nx,cos nx are all orthogonal to each
other. These frequencies are called harmonics.
This idea is key to understanding OFDM. The orthogonality allows simultaneous
transmission on a lot of sub-carriers in a tight frequency space without
interference from each other. In essence this is similar to CDMA, where codes are

8|Page
used to make data sequences independent (also orthogonal) which allows many
independent users to transmit in same space successfully.

Designing OFDM Transmitter & Receiver:

Consider that we want to send the following data bits using OFDM : D =
{d0,d1,d2,). The first thing that should be considered in designing the OFDM
transmitter is the number of subcarriers required to send the given data. As a
generic case, lets assume that we have N subcarriers. Each subcarriers are
centered at frequencies that are orthogonal to each other (usually multiples of
frequencies).

The second design parameter could be the modulation format that we wish to
use. An OFDM signal can be constructed using anyone of the following digital
modulation techniques namely BPSK, QPSK, QAM etc..,
The data (D) has to be first converted from serial stream into parallel stream
depending on the number of sub-carriers (N). Since we assumed that there are N
subcarriers allowed for the OFDM transmission, we name the subcarriers from 0
to N-1. Now, the Serial to Parallel converter takes the serial stream of input bits
and outputs N parallel streams (indexed from 0 to N-1). These parallel streams are
individually converted into the required digital modulation format (BPSK, QPSK,
QAM etc..,). Lets call this output S0,S1,..SN. The conversion of parallel data (D) into
the digitally modulated data (S) is usually achieved by a constellation mapper,
which is essentially a look up table (LUT). Once the data bits are converted to
required modulation format, they need to be superimposed on the required
orthogonal subcarriers for transmission. This is achieved by a series of N parallel
sinusoidal oscillators tuned to N orthogonal frequencies (f0,f1,fN-1). Finally, the
resultant output from the N parallel arms are summed up together to produce
the OFDM signal.

Since the OFDM signal is in time domain, IFFT is the appropriate choice to use in
the transmitter, which can be thought of as converting frequency domain samples
to time domain samples. Well, you might ask : s(t) is not in frequency domain and
they are already in time domain; so whats the need to convert it into time domain
again ? The answer is IFFT/FFT equation comes handy in implementing the
conversion process and we can eliminate the individual sinusoidal multipliers

9|Page
required in the transmitter/receiver side. The following figure illustrates, how the
use of IFFT in the transmitter eliminates the need for separate sinusoidal
converters. Always remember that IFFT and FFT blocks in the transmitter are
interchangeable as long as their duals are used in receiver.

An OFDM system is defined by IFFT/FFT length N ,the underlying modulation


technique ( BPSK/QPSK/QAM), supported data rate, etc..,. The FFT/IFFT length N
defines the number of total subcarriers present in the OFDM system. For
example, an OFDM system with N=64 provides 64 subcarriers. In reality, not all
the subcarriers are utilized for data transmission. Some subcarriers are reserved
for pilot carriers (used for channel estimation/equalization and to combat
magnitude and phase errors in the receiver) and some are left unused to act as
guard band. OFDM system do not transmit any data on the subcarriers that are
near the two ends of the transmission band ( Not necessarily at the ends of the
bands, implementation may differ). These subcarriers are collectively called guard
band. The reservation of subcarriers to guard bands helps to reduce the out of
band radiation and thus eases the requirements on transmitter front-end filters.
The subcarriers in the guard band are also called Null subcarriers or virtual
subcarriers.

10 | P a g e
Cyclic Prefix:

Consider a non-ideal channel h(t)(t), that causes delay dispersion. Delay


dispersion manifests itself as Inter Symbol Interference (ISI)on each subcarrier
channel due to pulse overlapping. It will also cause ICI (Inter Carrier Interference )
due to the non-orthogonality of the received signal. Adding cyclic prefix to each
OFDM symbol mitigates the problems of ISI and ICC by removing them altogether.

Lets say, without cyclic prefix we transmit the following N values (N=Nfft=length
of FFT/IFFT) for a single OFDM symbol.

X0,X1,X2,,XN1X0,X1,X2,,XN1

Lets consider a cyclic prefix of length Ncp, ( where Ncp<N ), is formed by copying
the last Ncp values from the above vector of X and adding those Ncp values to the
front part of the same X vector. With a cyclic prefix length Ncp, ( where Ncp<N ),
the following values constitute a single OFDM symbol :

11 | P a g e
The key ideas behind adding cyclic prefix :

1) Convert linear convolution in to circular convolution which eases the


process of detecting the received signal by using a simple single tap
equalizer

2)
2) Help combat ISI and ICC.

When a cyclic prefix of length Ncp is added to the OFDM symbol, the output of
the channel (r) is given by circular convolution of channel impulse response (h)
and the OFDM symbols with cyclic prefix (x).

r=hxr=hx

As we know, for the discrete signals, circular convolution in the time domain
translates to multiplication in the frequency domain.Thus, in frequency domain,
the above equation translates to

R=HXR=HX

At the receiver, R is the received signal (in Frequency domain) and our goal is to
estimate the transmitted signal (X) from the received signal R. From the above
equation, the problem of detecting the transmitted signal at the receiver side
translate to a simple equalization equation as follows
12 | P a g e
X^=RHX^=RH

After the FFT performed at the receiver side (i.e. after the FFT block in the
receiver side), a single tap equalizer (which essentially implements the above
equation) is used to estimate the transmitted OFDM symbol. It also corrects the
phase and equalizes the amplitude.

A basic OFDM architecture with Cyclic Prefix is given below. (In the following
diagram, symbols represented by small case letters are assumed to be in time
domain, whereas the symbols represented by uppercase letters are assumed to
be in frequency domain).

ADVANTAGES OF OFDM

High spectral efficiency as compared to other double sideband modulation


schemes, spread spectrum.

Can easily adapt to severe channel conditions without complex time-domain


equalization.

Robust against narrow-band co-channel interference

Robust against intersymbol interference (ISI) and fading caused by multipath


propagation

Efficient implementation using fast Fourier transform

Low sensitivity to time synchronization errors

Tuned sub-channel receiver filters are not required (unlike conventional FDM)

Facilitates single frequency networks (SFNs) (i.e. transmitter macrodiversity)

13 | P a g e
Resilient to narrow-band effects :- This advantage of OFDM deals with
Channel coding and interleaving
OFDM is used in conjunction with channel coding and almost always
uses frequency and/or time interleaving.
Frequency (subcarrier) interleaving increases resistance to frequency
selective channel conditions such as fading.
Time interleaving ensures that bits that are originally close together
in the bit-stream are transmitted far apart in time, thus mitigating
against severe fading.
A classical type of error correction coding used with OFDM-based
systems is convolutional coding.
Immunity to selective fading :- One of the main advantages of
OFDM is that is more resistant to frequency selective fading than
single carrier systems because it divides the overall channel into
multiple narrowband signals that are affected individually as flat
fading sub-channels.

DISADVANTAGES OF OFDM

The drawbacks of OFDM are very few in comparison to the advantages of OFDM.
The main drawback of OFDM is PAPR which is discussed in point (2).
The disadvantages of OFDM are as follows :-
1) Sensitive to carrier offset and drift :- A disadvantage of OFDM is that is
sensitive to carrier frequency offset and drift. Single carrier systems are less
sensitive.

14 | P a g e
2) High peak to average power ratio (PAPR) :- The most important disadvantage
of OFDM is high PAPR An OFDM signal has a noise like amplitude variation
and has a relatively large dynamic range, or peak to average power ratio. This
impacts the RF amplifier efficiency as the amplifiers need to be linear and
accommodate the large amplitude variations and these factors mean the
amplifier cannot operate with a high efficiency level.

PAPR is the most vital disadvantage of OFDM. Hence reducing PAPR is immensely
important for efficient utilisation & application of OFDM. Thus a broad &
elaborate discussion is required about the factors on which PAPR depends & how
to effectively reduce & control PAPR and the discussion is done in the following
pages.

15 | P a g e
WHAT IS PAPR

PAPR:- A major disadvantage that arises in multicarrier systems like OFDM is the
resulting non-constant envelope with high peaks [27]. When the independently
modulated sub-carriers are added coherently, the instantaneous power will be
more than the average power. Consider the OFDM signal x(t) dened as
x(t)=exp(j2fct)a(t) , (fc is the lowest sub-carrier frequency), where N subcarriers
are added together. If N is large enough, then, based on central-limit theorem
(CLT), the resulting signal x(t) will be close to a complex Gaussian process [28].
This means that both of its real and imaginary parts are Gaussian distributed and
its envelope and power follows Rayleigh and exponential distributions
respectively. The PAPR for the continuous-time signal x(t) is the ratio of the
maximum instantaneous power to the average power. For the discrete-time
version x[n], PAPR is expressed as

where E[.] is the expectation operator. It is worth mentioning here that PAPR is
evaluated per OFDM symbol. Figure 3 illustrates how a high peak is obtained by
adding four sinusoidal signals with different frequencies and phase shifts
coherently. The resulting signals envelope exhibits high peaks when the
instantaneous amplitudes of the different signals have high peaks aligned at the
same time. Such high peaks will produce signal excursions into nonlinear region
of operation of the power amplier (PA) at the transmitter, thereby leading to
non linear distortions and spectral spreading. Since IFFT is used to generate the
FDM signal, the resulting discrete-time OFDM signal samples are obtained at the
Nyquist-rate. The peak value computed using these samples may not coincide

16 | P a g e
with the peak value of the continuous-time OFDM signal [30]. Hence,
oversampling by a factor greater than 1 is used to increase the accuracy. It is
found that the PAPR of the oversampled discrete-time signal offers an accurate
approximation of the PAPR of the continuous-time OFDM signal if the
oversampling factor is at least 4

17 | P a g e
FACTORS ON WHICH PAPR DEPENDS

The following factors are used to reduce & control the high Peak to Average
Power Ratio (PAPR)

1) Bit Error Rate :- The performance of a modulation technique can be


quantied in terms of the required signal-to-noise ratio (SNR) to achieve a
specic bit error rate (BER). Although the main focus of PAPR reduction
techniques is to reduce the CCDF, this is usually achieved at the expense of
increasing the BER. Clipping the high peaks of the OFDM signal by the PA
causes a substantial in-band distortion that leads to higher BER. Other
techniques may require that side information be transmitted as well. If the
side information is received incorrectly at the receiver, the whole OFDM
symbol is recovered in error and the BER performance degrades.

2) Transmitted Signal Power :- Some PAPR reduction techniques require that the
average power of the transmitted signal be increased. If the linear region of
the PA is not stretched to accommodate the new signal, the signal will
traverse the nonlinear region leading to higher distortions and degraded BER
performance. However, this solution increases the hardware cost

3) Spectral Spreading :- Due to the limit imposed on the maximum peak of the
OFDM signal by the PA, an increase is encountered in both the in-band and
out-of-band distortions. The second causes undesirable increase in the power
of the side lobes of the power spectral density (PSD) of the OFDM signal. This
effect is referred to as spectral spreading or spectral regrowth. As
demonstrated in Fig. 5, when the nonlinearity of the PA is higher, IBO is
smaller, and the spectral spreading is higher. Spectral spreading leads to
higher interference between the sub-bands of the OFDM signal, unless the

18 | P a g e
frequency separation between adjacent sub-carriers is also increased to
maintain orthogonality. However, this solution has the disadvantage of
lowering the spectral efciency.

4) Computational Complexity :- Generally, techniques with increased complexity


have better PAPR reduction capability with less undesirable effects than
simple ones. However, complex techniques require additional hardware,
processing power and time. In practice, both hardware and processing
complexity should be as minimum as possible to support real-time system
operations and minimize cost.

5) Data Rate Loss :- Some PAPR reduction techniques cause some data rate loss
due to extra bandwidth required to send side information. Other techniques
may require some non-information symbols to be dedicated for controlling
PAPR. If the information data rate is required to be the same as that prior to
applying the technique, a bandwidth expansion will be a direct result.

6) Other Factors :- PAPR reduction techniques should take into consideration the
effect of nonlinear devices in the transmitter such as the DAC, mixer, transmit
lter and PA. The nonlinearity introduced by these devices and their cost are
two important factors in the system design process.

19 | P a g e
PAPR REDUCTION TECHNIQUES

A large PAPR would drive PAs at the transmitter into saturation, producing
interference among the subcarriers that degrades the BER performance and
corrupts the spectrum of the signal. To avoid driving the PA into saturation, the
average power of the signal may be reduced. However, this solution reduces the
signal-to-noise ratio and, consequently, the BER performance. Therefore, it is
preferable to solve the problem of high PAPR by reducing the peak power of the
signal.
Many PAPR reduction techniques have been proposed in the literature.
These techniques can be broadly classied into three main categories:-
1) Signal distortion techniques,
2) Multiple signaling and probabilistic techniques.
3) Coding techniques.

The PAPR reduction techniques can also be mainly classified into two basic types:-
1) Distortion based techniques:- This is the scheme that introduces
spectral re-growth belong to this category. Distortion based techniques
are the most straightforward PAPR reduction methods. Furthermore,
these techniques distort the spectrum, this spectrum distortion or
spectral re-growth can be corrected to a certain extent by using
filtering operation
2) Non-distortion based techniques :- These types of PAPR reduction
schemes do not distort the shape of the OFDM signal and therefore no
spectral re-growth take place
Distortion based techniques can be further sub-classified into the
following :-
a) CLIPPING AND FILTERING.
b) COMPANDING.
On the other hand Non-distortion based techniques can be further sub-
classified into the following :-

20 | P a g e
a) CODING SCHEME.
b) PARTIAL TRANSMIT SEQUENCES(PTS).

21 | P a g e
This represents the Taxonomy of PAPR reduction technique

22 | P a g e
A detailed discussion about the PAPR reduction techniques are done below :-
1) Signal Distortion Technique :- Signal distortion techniques reduce the PAPR
by distorting the transmitted OFDM signal before it passes through the PA.
The most well-known signal distortion techniques are
a) Clipping and ltering,
b) Peak windowing,
c) Companding,
d) Peak cancellation.

These techniques reduce the PAPR signicantly but they introduce both
in-band and out-of-band distortion, leading to increase in BER. Although
the OFDM transmitted signal may have a high PAPR, the high magnitude
peaks occur rarely and most of the signal power will depend on low
amplitude samples. Therefore, it is possible to remove the high peaks
without signicantly distorting the signal. Hence, PAPR may be reduced
at the expense of some tolerable increase in BER.

CLIPPING AND FILTERING:- Clipping process of PAPR reduction is a Distortion


Based Technique of PAPR reduction Clipping is one of the simplest techniques to
reduce the PAPR of OFDM signal. It reduces the peak of the OFDM signal by
clipping the signal to the desired level. This operation can be implemented on
discrete time samples prior to the DAC or by designing the power amplifiers with
saturation level lesser than the OFDM signal dynamic range. The amplitude
clipper limits the peak of the envelope of the input OFDM signal to a
predetermined threshold value ( ) or otherwise passes the input signal
unperturbed. The clipping operation can be mathematically defined as

where [ ] and [ ] are the clipped OFDM signal and the phase of ( ). It is a
nonlinear operation and therefore causes both in-band distortion and out-of-
band radiation. The in-band distortion is treated as noise and therefore results in

23 | P a g e
error performance degradation. If clipping level o is too small, then out-of-band
radiation becomes more significant. The outof-band radiation can be eliminated
by using filtering operation but filtering operation causes peak re-growth in OFDM
signal, and as a result the amplitude of time domain OFDM signal again exceeds
the clipping level ( o ). The BW efficiency of the OFDM system decreases due to
spectral re-growth. The computational complexity of the clipping scheme is found
to be the least in comparison to other distortion PAPR reduction schemes but at
the same time its BER performance is very poor.

Iterative Clipping and Filtering :- Jean Armstrong proposed the repeated clipping
and filtering scheme [25], in which clipping and frequency domain filtering
operations are repeated several times to reduce both the out of-band radiation
and PAPR to the desired level. The PAPR performance and amount of out- of-band
radiation mainly depends on the number of iterations to be performed, more the
number of iterations lesser is the value of out-of-band radiations and PAPR. But,
the computational complexity of iterative clipping and filtering scheme increases
with the number of iterations. We have studied the effect of iterative clipping
and filtering of OFDM signal through computer simulation. In this simulation, we
have considered an amplitude clipper with a threshold = 0.01 and four
iterations. The time domain and frequency domain OFDM signals are shown after
each iteration. It is observed from following graph that the original time domain
OFDM signal has large envelope fluctuation and has a PAPR of 3.425dB.

Therefore, in order to reduce the PAPR, the OFDM signal is passed through a
clipper with a threshold = 0.01 , as a result the PAPR of the OFDM signal
reduced to 1.396dB, but due this nonlinear operation, both in-band distortion and
out-of-band radiation take place as shown in Fig. 3.6. The frequency domain
filtering is used to eliminate the out-of-band radiation but the peaks of the time
domain OFDM signal re-grows after performing the filtering operation. Thus the
PAPR of OFDM signal again gets increased to 1.746dB. In order to further limit the
PAPR and out-of-band radiation the clipping and filtering operation are required
to be performed many times to achieve the desired value of PAPR and out-of-
band radiation.

24 | P a g e
For a large number of iterations the computational complexity of the system gets
increased, which makes the system slow. One of the main drawbacks of this
technique is in-band distortion, which canNOT be eliminated by frequency
domain filtering operation. Therefore, degradation in error performance of OFDM
system takes place.
PEAK WINDOWING :- Unlike peak clipping where the peaks that exceed a
predetermined threshold are hard-limited, peak windowing limits such high peaks
by multiplying them by a weighting function called a window function. Many
window functions can be used in this process as long as they have good spectral
properties. The most commonly used window functions include Hamming,
Hanning and Kaiser windows. To reduce PAPR, a window function is aligned with
the signal samples in such a way that its valley is multiplied by the signal peaks
while its higher amplitudes are multiplied by lower amplitude signal samples
around the peaks. This action attenuates signal peaks in a much smoother way
compared to hard clipping, resulting in reduced distortion.
COMPANDING :- Companding transforms are typically applied to speech signals
to optimize the required number of bits per sample. Since OFDM and speech
signals behave similarly in the sense that high peaks occur infrequently, same
companding transforms can also be used to reduce the OFDM signals PAPR.
Besides having relatively low computational complexity compared to other PAPR
reduction techniques, companding complexity is not affected by the number of
subcarriers. Also, companding does not require side information and hence does
not reduce bit rate. Their simplicity of implementation and the advantages they
offer make companding transforms an attractive PAPR reduction technique. The
PAPR reduction obtained by companding transforms comes though with the price
of increasing the BER.
Companding is another popular PAPR reduction scheme. Companding is a
composite word formed by combining compressing and expanding. In this
scheme, at the transmitter a signal with high dynamic range is applied to a
compander and at the receiver a decompanding function (the inverse of
companding function) is used to recover the original signal. Initially, it was used in
digital communication systems to increase the dynamic range of digital to analog
converters (DACs). The -law and A-law are the two most popular compressing

25 | P a g e
functions used worldwide. Wang et al. proposed a scheme a -law companding
scheme to reduce the PAPR of OFDM signal. According to this scheme, the time
domain OFDM signal
( x[n] ) is applied to a -law compressor to produce a companded OFDM signal
given by

where, A is the peak value of the OFDM signal before companding and is the
parameter controlling nonlinearity of the companding function, sgn(.) and ln(.) are
the standard signum and natural logarithmic function respectively. The discrete
time domain companded OFDM signal is converted to analog signal using D/A
converter and then amplified using HPA to achieve the desired power level. The
amplified signal is then transmitted over the communication channel. At the
receiving end, the received signal ( r[n] ) is applied to the expander to recover the
transmitted time domain signal. The time domain OFDM signal at expander
output is given by

Here, the received signal ( r[n] ) is nothing but the companded OFDM signal (
corrupted by quantization and channel noise. Before companding, the real and
the imaginary parts of the complex time domain OFDM signal have Gaussian
distribution. But, after applying the -law companding scheme, the PDF of real
and imaginary parts no longer remains Gaussian distributed. Figure below shows

26 | P a g e
the probability distribution function of real part before and after companding of
OFDM signal.

PDF of real part of original and -law companded OFDM amplitude


It can be easily observed from the above figure that due to -law companding
operation the probability of real part of the OFDM signal with small amplitude
gets decreased whereas the probability of OFDM signal with large amplitude gets
increasedsignal . A similar phenomenon can be observed for the imaginary part of
the OFDM signal. Therefore, the average power of the OFDM signal increases. In
-law companding scheme the peak value of the OFDM signal before and after

27 | P a g e
companding remain same, which keeps peak power of the OFDM signal
unchanged. Therefore, -law companding scheme decreases the PAPR of the
OFDM signal. By controlling the parameter , the average power of the OFDM
signal can be controlled and hence the desired value of PAPR can be obtained.
The overall PAPR reduction capability and BER performances of -law
companding scheme is found to be better than clipping and filtering scheme.
Due to increase in the average power level of the OFDM signal, the overall error
performance of the system degrades in comparison to OFDM signal without
companding; higher the value of the , better is the PAPR reduction, but at the
same time, error performance degradation is also higher.

The figure below shows the CCDF of PAPR for normal and -law companded
OFDM signal. In this simulation, we have considered a QPSK modulated OFDM
system with N=256 subcarriers. We have used =254 for companding the
OFDM signal. It can be observed from figure below that -law companding

28 | P a g e
scheme, achieves a PAPR reduction capability of 5.2dB for a CCDF of 10^(-4).

Comparison of PAPR performance of normal and -law companded OFDM signal

BER performance of the normal and -law companded OFDM signal is shown in
fig. below. The BER performance of OFDM system utilizing -law companding is
worse than the normal OFDM signal. Normal OFDM system requires 4dB less SNR
to achieve a BER of 10^(-5).

29 | P a g e
BER performance comparison of Normal OFDM and -law companding

All of the techniques discussed in this section do not require any side information
to recover the data signal at the receiver and hence the data rate of the system
remains same even after applying PAPR reduction scheme. The computational
complexity of these schemes is also very less. However, the price paid for using a
distortion technique is in-band distortion, out of-band radiation and ICI, which
adversely affects the bit error rate (BER) of the system.
PAPR Reduction using Quadrilateral Companding Transform:- A quadrilateral
companding transform is proposed to reduce the PAPR of the OFDM signal. The
mathematical expression of companding function is derived by transforming the
distribution of original OFDM signal to a quadrilateral distribution. The
mathematical expression of de-companding function is also derived which is used
at the receiver to retrieve the original OFDM signal. The proposed scheme is the
most generalized companding transform because the existing exponential
companding transform, trapezoidal companding and trapezium distribution based
30 | P a g e
companding methods are special cases of the suggested quadrilateral
companding transform. The PAPR and BER performance of the proposed scheme
is compared with other existing companding transforms.

PEAK CANCELLATION : - In this technique, a peak cancellation waveform is


appropriately generated, scaled, shifted and subtracted from the OFDM signal at
those segments that exhibit high peaks. The generated waveform is band limited
to certain peak cancellation tones that are not used to transmit data. Peak
cancellation can be carried out after the IFFT block of the OFDM transmitter as
shown in Figure 1 below by subtracting the peak cancellation waveform from the
OFDM signal whenever a potential peak higher than a certain threshold is
detected. Fig. 2(a) illustrates a segment of an OFDM signal with 256 subcarriers.
Figure 2(b) shows the detected potential peaks of the signal and Fig. 2(c) shows a
randomly chosen sinc function as the peak cancellation waveform. While
performing the peak cancellation process, care should be taken not to create new
peaks

Fig. 1) Block Diagram for Peak cancellation in OFDM transmitter :-

31 | P a g e
Fig. 2) Peak Cancellation; (a) OFDM signal, (b) identied peaks, (c) scaled
& shifted peak.

2) Multiple Signaling and Probabilistic Techniques :- These techniques work in


one of two ways. One way is to generate multiple permutations of the OFDM
signal and transmit the one with minimum PAPR. The other way is to modify the
OFDM signal by introducing phase shifts, adding peak reduction carriers, or
changing constellation points. The modication parameters are optimized to
minimize PAPR.The most well-known Multiple Signaling and Probabilistic
Techniques are

a) Active Constellation Extension(ACE)


b) Tone Injection (TI) & Tone Reservation (TR)
c) Selective Mapping(SLM)
d) Partial Transmit Sequence(PTS).

32 | P a g e
Now let us broadly discuss each of these techniques
ACTIVE CONSTELLATION EXTENSION (ACE) :- This is a non-distortion technique of
PAPR reduction technique. ACE scheme is first proposed by Krongold and Jones in
2003 to reduce the PAPR of the OFDM signal. Its operation is very similar to tone
injection scheme for PAPR reduction [38]. In this scheme, the outer constellation
points are extended towards outer side of signal constellation with an aim to
cancel the peak of the resultant time domain OFDM signal. Figures below show
the principles of ACE scheme for an OFDM signal with QPSK and 16-QAM
modulation in each subcarrier. It can be seen from Figure below that any of the
four points of QPSK constellation can be extended outwards in such a way that
the minimum Euclidean distance between two points does not reduce. In case of
QPSK constellation the extended points remains in the same quadrant as before.
As shown in Figure below the outer 12 points of 16-QAM constellation can be
extended towards outer side of the original constellation for reducing the peak of
the OFDM signal. Here, the most important part is to find the extended points for
minimizing the peak of the OFDM signal. One can utilize POCS or smart gradient
project (SGP) scheme for optimization of the same. The POCS algorithm has slow
convergence but provides the optimal solution to the problem, whereas SGP
provides a suboptimal solution with faster convergence.

ACE constellation for (a) QPSK (b) 16-QAM constellation

33 | P a g e
ACE scheme is free from data rate loss and SI; but due to constellation size
extension, the average transmitted power of the OFDM signal may increase,
which restricts its use for large constellations.
TONE RESERVATION & TONE INJECTION :- This is a distortionless PAPR reduction
technique. Tone reservation was first described by Tellado and Cioffi in 1997. The
basic idea is to reserve certain numbers (R) of subcarriers or tones for reducing
the PAPR of the OFDM signal. The subcarriers or tones reserved for PAPR
reduction are known as PAPR reduction tones (PRT). Here, the main objective is
to choose the value of data signals to be transmitted on reserved tones to
minimize the PAPR of time domain OFDM signal. The PAPR reduction capability of
this scheme mainly depends on the number of PRTs and their location in the
frequency band. The PRT locations are known in advance to both transmitter and
receiver. The PRTs are not used for data transmission and therefore it results in
data rate loss. Let be the data
and PRT blocks, the PRT locations in Z and locations of subcarriers for data
transmission in X are chosen dis-jointly, such that their point wise multiplication
= 0 is Zero. Consider a tone reservation scheme with R PRTs located at
and the data symbols are transmitted only on the sub-carriers with index
and
After applying tone reservation scheme, the frequency domain OFDM signal
( ) is given by following expression

The corresponding time domain OFDM signal can


be found as

34 | P a g e
where x[n] and z[n] are the N point IFFTs of X and Z respectively. In order to
reduce the PAPR of the time domain OFDM signal, the peak is minimized by
finding the optimal value of PRTs i.e. . For this, one can utilize convex
optimization technique that can be cast as linear programming problem. To find
, projection on to convex set (POCS) or gradient search algorithm can be used .In
order to reduce data rate loss in TR scheme, tone injection (TI) scheme is
proposed. In this scheme, the PRTs and data symbols can be transmitted on the
same subcarriers. The basic idea of TI is to extend the constellation, such that one
constellation point may correspond to more than one point in the extended
constellation. In this scheme PRTs and data symbols are not mutually exclusive,
therefore at the receiver; some efficient scheme to separate the PRTs has to be
utilized. One of the solutions to this problem is to use PRTs of the type
where q is a positive integer and are chosen for reducing the PAPR
of OFDM signal. If we choose and then for q>1, each of the
M-PSK or M-QAM modulated symbol corresponds to 9 different symbols of
extended constellation, and therefore result in a constellation with M 9 points.
The expanded constellation symbol is given by

The average power of the signal gets increased. At the receiver, the effect of
can be eliminated by performing modulus q operation on received data symbols.
An example of expanded constellation obtained from TI scheme for QPSK-OFDM
system is shown in Figure below.
Tellado suggested to use a minimum value of where d is the
minimum distance between two constellation points. The extended constellation
points have spacing with 1 to avoid BER performance degradation.

35 | P a g e
Expanded 16-QAM constellation for Tone Injection (TI) technique

SELECTIVE MAPPING :- SLM is another popular distortion-less PAPR reduction


technique first described by Cimini et al. In this scheme, a set of U independent
data sequences is generated by multiplication of a block of N modulated data
symbols with a phase sequence set containing U phase vectors of
length N . The set of U independent signals can be mathematically expressed as
follows

Here, each phase vector has a length N and its


elements, i.e.
phase factors, where W denotes the
number of
phase factors. The set of U independent discrete-time OFDM signals

can be obtained from a set of U data sequences as

36 | P a g e
Out of these U signals, one of them satisfying following criterion is selected for
transmission

where is the index of phase vector in the phase sequence set


that
generates an OFDM signal with least PAPR. Therefore, the information about
should be transmitted along with each OFDM symbol, which requires
bits, if straight binary coding is used to encode the SI. The SI bits are extremely
important for data recovery at the receiver; therefore sometimes we sacrifice the
data rate and allocate few redundant bits to ensure accurate recovery of
. But it further increases the loss in data rate of a SLM-OFDM system.

37 | P a g e
Block diagram of conventional SLM-OFDM transmitter
Figure given below shows comparison of the PAPR performance of
SLM-OFDM system for various values of U.

PAPR performance comparison of SLM-OFDM system for various


values of U

38 | P a g e
In this simulation, we have considered a QPSK modulated OFDM signal with N=64
subcarriers. It can be seen from the above figure that the PAPR reduction
capability of SLM-OFDM system increases by increasing the value of the number
of phase sequences (U) in phase sequence set. But, this gain in PAPR reduction
capability is obtained at the cost of increased computational complexity because
for a phase sequence set with U phase sequences requires U IFFT operations
tofind the OFDM signal with lowest possible PAPR
( ). Moreover, higher the value of U, more is the number of bits required to
encode the side information and therefore data rate loss is also more.
PARTIAL TRANSMIT SEQUENCE:- PTS is one of the most popular distortion-less
PAPR reduction scheme. In this scheme a block of N modulated data symbols
is partitioned into S disjoint sub-blocks, where S <N . After partitioning, S data
sub-blocks are represented by
here the length of each data sub-block is N and all of them are disjoint in a sense
that the value of is non-zero only for one particular value of
,
therefore we have

After this, IFFT of each of the data sub-block is taken to obtain S partial transmit
sequences
given by

39 | P a g e
The partial transmit sequences are multiplied by phase rotation factors b(s) and
all of them are combined to obtain a time domain OFDM signal ( x ) given by
following expression

where b(s) is the phase rotation factor for data sub-block. Here, the
objective of combining the partial transmit sequences after multiplication with
phase factor is to obtain a time domain OFDM signal ( x) with lowest possible
PAPR. Therefore, to find the optimal values of phase factors to achieve lowest
possible PAPR of OFDM signal x, following optimization criterion is used

where are the optimized phase rotation factors for sub-blocks


respectively.
The partitioning schemes for PTS based PAPR reduction schemes are majorly classified into (i)
adjacent , (ii) interleaved , (iii) pseudorandom partitioning. It has been reported that PAPR
reduction capability of PTS-OFDM system using pseudorandom partitioning is better than that
of adjacent and interleaved partitioning schemes.

40 | P a g e
Block diagram of conventional PTS-OFDM system transmitter
The PAPR reduction capability of PTS-OFDM system increases by increasing
number of partitions (S). But, in this scheme for S data sub-blocks, S IFFT
operations are required to calculate X which results in high computational
complexity. Therefore, the number of partitions (S) is restricted to 4.
The information about the phase rotation factors
used at the transmitter for PAPR reduction is required to be sent along with each
OFDM symbol for data recovery at the receiver. The information used for this
purpose is called side information (SI), which reduces the effective data rate of
PTS-OFDM system. If straight
binary coding scheme is used then bits per OFDM symbol are
required to encode the SI . The loss in data rate will further increase if any error
control coding with low code rate is used for encoding the SI.

The figure below shows the PAPR performances of OFDM signal without PAPR
reduction , PTSOFDM system with 4 partitions and two phase factors (1,-1), and
PTS-OFDM system with 4 partitions and four phase factors (1, j,-1, -j). In this

41 | P a g e
simulation we have considered a QPSK modulated OFDM system with N=256 sub-
carriers.

Comparison of PAPR performance of PTSOFDM with W=2 and 4


The PAPR reduction capability of the PTS increases with the increase in the value
of W. PTS with W=2 and W=4, achieve a PAPR reduction capabilities of 2.5dB and
3.5 dB respectively, for a CCDF of , over OFDM signal without PAPR
reduction.
The PAPR reduction capability of PTS-OFDM system with W=4 is more than that
of PTS-OFDM system with W=2 because for M=W=4 there are 64 alternative
signals, whereas for M=4 and W=2 there are only 8 alternative signals. More is
the number of alternative signals more is the PAPR reduction.

42 | P a g e
3) Coding Schemes :- Coding scheme can be used to reduce the PAPR of the
OFDM signal. Consider a BPSK modulated OFDM system with N subcarriers and a
coding scheme with code rate 1/2. The code rate 1/2 encoder converts N input
bits to a block of 2N bits.
Therefore, there are possible combinations of 2N bit long code words, out
of which
code words with lowest PAPR are selected for transmission of N
information bits. This scheme can effectively reduce the PAPR of the OFDM signal,
but it requires an exhaustive
search to find low PAPR code words out of possible code
words. For large number of subcarriers, it requires large memory blocks to store
lookup table for encoding and decoding operations. It also results in data rate loss
because for every N information bits, we have to transmit 2N number of encoded
bits. Example: Consider a BPSK modulated OFDM system with four subcarriers
and code rate 3/4 encoder. In this case there are 16 possible combinations of data
blocks; and PAPR of each possible data block is calculated using the formula given
below,

and The values are shown in Table


below

Data Block PAPR(dB) Data Block PAPR(db)

1 1 1 1 6.0 -1 1 1 1 2.3

1 1 1 -1 2.3 -1 1 1 -1 3.7

1 1 -1 1 2.3 -1 1 -1 1 6.0

43 | P a g e
1 1 -1 -1 3.7 -1 1 -1 -1 2.3

1 -1 1 1 2.3 -1 -1 1 1 3.7

1 -1 1 -1 6.0 -1 -1 1 -1 2.3

1 -1 -1 1 3.7 -1 -1 -1 1 2.3

1 -1 -1 -1 2.3 -1 -1 -1 -1 6.0

There are three main types of distortion-less coding schemes PAPR reduction
techniques which are listed below

a) Linear Block Code Scheme


b) Turbo Code Scheme
c) Golay Code scheme
These types of PAPR reduction schemes do not distort the shape of the OFDM
signal and therefore no spectral re-growth take place and hence are called non
distortion PAPR reduction technique.
LINEAR BLOCK CODE SCHEME :- Instead of dedicating some bits of the code word
to enhance BER performance, these bits are now dedicated to reduce PAPR. The
goal is to choose the code words with low PAPR for transmission. A simple linear
block coding (LBC) scheme was proposed where 3 bits are mapped into 4 bits by
adding a parity bit. A simple rate 3/4 cyclic code is used for any number of
subcarriers that is a multiple of 4 to reduce PAPR by more than 3 dB. Similar
performance with less complexity was obtained in using the proposed sub-block
coding (SBC) scheme, where long information sequences are divided into sub-
blocks, and an odd parity bit is added to each sub-block. The position of the added
parity bit is optimized to further reduce PAPR. Moreover, instead of one coding
scheme, two coding schemes can be used to encode each sub-block, and the
combination of the coded sub-blocks can be optimized to lower PAPR. Both
methods require the transmission of side information to the receiver to indicate
the locations of the parity bits or the coding schemes used to encode each sub-
block. This means that the bandwidth efciency of the system will be reduced. A

44 | P a g e
combined (8,4) LBC is used to provide error control capability and reduce PAPR of
a multicarrier modulation by 4 dB. Another simple LBC is proposed based on the
observation that regardless of the number of subcarriers, code words with equal
odd and even bit values have high PAPR. Therefore, PAPR can be reduced easily
by eliminating these code words by adding a simple bit code. It was proposed that
a rate-1/2 linear coding scheme to achieve the minimum PAPR and provide error
correction capability for an OFDM system utilizing QPSK modulation and four
subcarriers. The idea is to use a generator matrix to encode 4 bits into code words
of 8 bits (rate 1/2), and then multiply the resulting 8 bits by a phase rotator vector
of length 8 to have the encoded code words. Both the generator matrix and the
phase rotator vector are chosen to reduce PAPR. Similar precoding schemes using
proper generator matrices to reduce the PAPR of OFDM signals has been
presented in the literature. However, precoding schemes increase the error
probability of OFDM systems. By assuming that the transmitter has the
knowledge of the channel information and the receiver lter, the author
proposed a systematic procedure to choose a precoding matrix for achieving the
required PAPR reduction performance at the minimum error probability over
AWGN channels.

TURBO CODE SCHEME :- One way to exploit turbo codes for PAPR reduction is to
implement the SLM approach with the candidates generated by a turbo encoder
with various interleavers. The selected one for transmission is the one with the
least PAPR. Since no side information is required, this method avoids the BER
performance degradation resulting from incorrect recovery of side information in
conventional SLM approaches. Beside PAPR reduction benet, it is possible to
make use of the error correction capabilities of Turbo coding.
Another approach was proposed based on the dual bose-ray-chaudhuri (BCH)
codes. Dual BCH code has some appealing PAPR properties. Specically, the IFFT
of the code words of this code exhibits low envelope uctuations and

45 | P a g e
therefore low PAPR. However, the potential of this code for PAPR
reduction is restricted due to the lack of a practical decoder and the
large performance gap to the Shannon limit. To solve these problems,
the proposed scheme constructed a new code with favourable PAPR
properties based on dual BCH codes, exploited this code in a turbo
structure to obtain an adequately low BER and reduce the gap between
the performance and Shannon limit, and developed the associated
decoder based on the maximum a posteriori (MAP) criterion. The
bounded PAPR of the coded OFDM symbol using this scheme is
guaranteed. Also, it was shown that among the dual BCH codes only
those with one bit error correction capability have a PAPR considerably
lower than that of the other error correcting codes.
GOLAY CODE SCHEME :- Golay Complementary Sequences in Golay Code Scheme
can be used as code words to modulate the subcarriers of the OFDM system,
yielding a signal of PAPR with an upper bound of 2. Golay Complementary
Sequences satisfy the property that their out of phase autocorrelation function is
zero.
Let be a bipolar sequence such that
.
Dene the aperiodic autocorrelation function of the sequence a as follows

Let b be another bipolar sequence similar to a. Then the pair (a,b) is called a Golay
Complementary Pair if it satises

46 | P a g e
Each member in this pair is called Golay complementary sequence (GCS). Large
set of binary length Golay pairs can be found from certain second-order
cosets of the rst order Reed Muller codes . Schemes that combine the block
coding approach and the use of GCSs provide a powerful way for incorporating
both the capabilities of error correction and control over PAPR. However, the
usefulness of such approach is limited to OFDM systems with small number of
subcarriers. For OFDM systems with large number of subcarriers, this approach
results in transmission rate loss and increased computational complexity due to
the exhaustive nature of the search required to nd good codes.
Several recent papers studied the use of Golay sequences and Golay
complementary pairs with different constellations to reduce PAPR. The
construction of Golay sequences based on STAR-16-QAM constellation is studied
and an upper bound for PAPR is derived and found to be 2. The construction of
STAR-16-QAM constellation is done using scaled set sum of two QPSK
constellations. Another constellation, called the Asterisk 16-QAM is proposed in
to reduce PAPR when the data is encoded using Golay sequences and Golay
complementary pairs. This constellation maintains the same PAPR upper bound
of the STAR-16-QAM while enhancing the mean symbol error rate. A generalized
paradigm for the Asterisk 16-QAM constellation is presented where a new
constellation family controlled by a single parameter is proposed. Golay
sequences are used as the building blocks to construct a new family of 64-QAM
sequences that are not necessarily Golay sequences and that could outperform
other existing OFDM sequences in terms of both PAPR and code rate.

47 | P a g e
MATLAB CODE FOR PAPR REDUCTION
In this project we have presented a MATLAB CODE for reducing the Peak to
Average Power Ratio (PAPR) of a OFDM signal using the CLIPPING & FILTERING
Technique. The MATLAB code has executed successfully & we have also provided
the Graphs & Figures that we have obtained by executing the code.

clear all
clc
close
% A: Setting Parameters
M = 4; % QPSK signal constellation
no_of_data_points = 128; % have 128 data points
block_size = 8; % size of each ofdm block
cp_len = ceil(0.1*block_size); % length of cyclic prefix
no_of_ifft_points = block_size; % 128 points for the FFT/IFFT
no_of_fft_points = block_size
data_source = randsrc(1, no_of_data_points, 0:M-1);
figure(1)
stem(data_source); grid on; xlabel('Data Points'); ylabel('transmitted data phase
representation')title('Transmitted Data "O"')
qpsk_modulated_data = pskmod(data_source, M);
stem(qpsk_modulated_data);
scatterplot(qpsk_modulated_data);
title('MODULATED TRANSMITTED DATA');
num_cols=length(qpsk_modulated_data)/block_size;
data_matrix = reshape(qpsk_modulated_data, block_size, num_cols);
cp_start = block_size-cp_len;
cp_end = block_size;

48 | P a g e
for i=1:num_cols,
ifft_data_matrix(:,i) = ifft((data_matrix(:,i)),no_of_ifft_points);
for j=1:cp_len,
actual_cp(j,i) = ifft_data_matrix(j+cp_start,i);
end
ifft_data(:,i) = vertcat(actual_cp(:,i),ifft_data_matrix(:,i));
end
[rows_ifft_data cols_ifft_data]=size(ifft_data);
len_ofdm_data = rows_ifft_data*cols_ifft_data;
ofdm_signal = reshape(ifft_data, 1, len_ofdm_data);
figure(3)
plot(real(ofdm_signal));
xlabel('Time');
ylabel('Amplitude');
title('OFDM Signal');
grid on;
avg=0.4;
clipped=ofdm_signal;
for i=1:length(clipped)
if clipped(i) > avg
clipped(i) = avg;
end
if clipped(i) < -avg
clipped(i) = -avg;
end

49 | P a g e
end
figure(4)
plot(real(clipped));
xlabel('Time');
ylabel('Amplitude');
title('clipped Signal');
grid on;
noise = randn(1,len_ofdm_data) + sqrt(-1)*randn(1,len_ofdm_data);
without clipping
for i=1:length(ofdm_signal)
if ofdm_signal(i) > avg
ofdm_signal(i) = ofdm_signal(i)+noise(i);
end
if ofdm_signal(i) < -avg
ofdm_signal(i) = ofdm_signal(i)+noise(i);
end
end
figure(5)
plot(real(ofdm_signal));
xlabel('Time');
ylabel('Amplitude');
title('OFDM Signal after HPA');
grid on;
with clipping
avg=0.4;
for i=1:length(clipped)
if clipped(i) > avg

50 | P a g e
clipped(i) = clipped(i)+noise(i);
end
if clipped(i) < -avg
clipped(i) = clipped(i)+noise(i);

end
end
figure(6)
plot(real(clipped));
xlabel('Time');
ylabel('Amplitude');
title('clipped Signal after HPA');
grid on;
channel = randn(1,block_size) + sqrt(-1)*randn(1,block_size)
after_channel = filter(channel, 1, ofdm_signal);
awgn_noise = awgn(zeros(1,length(after_channel)),0);
recvd_signal = awgn_noise+after_channel;
recvd_signal_matrix = reshape(recvd_signal,rows_ifft_data, cols_ifft_data);
recvd_signal_matrix(1:cp_len,:)=[];
for i=1:cols_ifft_data,
fft_data_matrix(:,i) = fft(recvd_signal_matrix(:,i),no_of_fft_points);
end
recvd_serial_data = reshape(fft_data_matrix, 1,(block_size*num_cols));
qpsk_demodulated_data = pskdemod(recvd_serial_data,M);

figure(7)

51 | P a g e
stem(qpsk_demodulated_data,'rx');
grid on;

Page :- 42

xlabel('Data Points');
ylabel('received data phase representation');
title('Received Data "X"')
after_channel = filter(channel, 1, clipped);
awgn_noise = awgn(zeros(1,length(after_channel)),0);
recvd_signal = awgn_noise+after_channel;
Convert Data back to "parallel" form to perform FFT
recvd_signal_matrix = reshape(recvd_signal,rows_ifft_data, cols_ifft_data);
recvd_signal_matrix(1:cp_len,:)=[];
for i=1:cols_ifft_data,

fft_data_matrix(:,i) = fft(recvd_signal_matrix(:,i),no_of_fft_points);
end
recvd_serial_data = reshape(fft_data_matrix, 1,(block_size*num_cols));
Demodulate the data
qpsk_demodulated_data = pskdemod(recvd_serial_data,M);
figure(8)
stem(qpsk_demodulated_data,'rx');
grid on;
xlabel('Data Points');
ylabel('received data phase representation');
title('Received Data clipped "X"')

52 | P a g e
OUTPUT

53 | P a g e
54 | P a g e
55 | P a g e
56 | P a g e
RESULT AND DISCUSSION
OFDM is an efcient multicarrier modulation technique for both wired and
wireless applications due to its high data rates, robustness to multipath fading
and spectral efciency. Despite these advantages, it has the major drawback of
generating high PAPR, which drives the transmitters PA into saturation, causing
nonlinear distortions and spectral spreading. The literature is rich with PAPR
reduction techniques, which decrease PAPR substantially at the expense of
increased BER, increased transmitted power, reduced bit rate, or increased
complexity. This survey has discussed many important aspects of PAPR reduction
techniques and the impact of these techniques on a number of critical design
factors. Some absolutely essential mathematical formulations were presented
including the statistics of PAPR and the distribution of the OFDM signal. We
demonstrated that no single technique is the best under all circumstances and the
proper technique should be selected based on system requirements and available
resources. For example, in OFDM systems with a large number of subcarriers (N
256), signal distortion techniques and specically clipping and ltering are the
least demanding in terms of computational complexity, while achieving good
PAPR reduction. The subject of PAPR reduction assumes increased importance
due to the fact that future wireless systems are likely to apply OFDM structures
with higher number of subcarriers than present ones in order to achieve higher
data rates and mobility. This implies that the problem of developing PAPR
reduction schemes for OFDM systems that are capable of mitigating the problem
with best performance trade-offs, including minimum complexity and cost, is a
rich subject with exciting possibilities for conducting further research. Besides
providing an extensive set of references to the subject of PAPR reduction
techniques, this survey brings up to date previously available surveys with a
treatment of most recent research as well as provides original contributions with
simulations, complexity analyses, and a treatment of the topic under transmitted
power constraint. The authors strongly believe that this survey will serve as a
valuable pedagogical resource to researchers, OFDM system architects, designers,
and developers by providing them an understanding of the current research
contributions in the area of PAPR reduction in OFDM systems, the different
available techniques and their trade-offs towards developing more efcient and
practical solutions.

57 | P a g e
Future work

The search for a good PAPR reduction and ICI cancellation scheme for OFDM
systems still faces many challenges. Future research could be carried out in the
following directions to further improve the PAPR reduction and ICI cancellation in
OFDM systems

Criterion for parameters selection to optimize both BER and PAPR


performances of QCT.
In this thesis we have derived the companding and de-companding
functions for QCT, which are functions of three independent parameters ( l
ba and , ). By taking few sets of these parameters as examples, we have
shown a trade-off between PAPR reduction capability and BER
performance, and proved the outperformance of the proposed schemes
over EC, TC and TDBC. However, there is still a need to find a criterion for
independent selection of parameters for joint optimization of PAPR and
BER performances.

Analytical results for BER performance of OFDM system using QCT over
AWGN and fading channels.
In this thesis, we have derived an analytical expression for PAPR
performance of QCT, whereas the BER performance of proposed
companding transform over AWGN and fading channels is evaluated by
computer simulations. QCT has a non-linear function and in mathematical
analysis of BER, the noise term cannot be separated from the desired signal
at the receiver. But the derivation of a formula for BER performance may
be possible by taking some approximations to separate the noise term from
desired signal term.
Three dimensional trade-off between PAPR, BER and spectral spreading.
In this thesis we have ignored the effect of spectral spreading because it
can be eliminated using iterative oversampling and filtering. If it is not
eliminated then its effect can be taken into consideration to get the

58 | P a g e
parameter of companding transform for achieving three dimensional
between PAPR, BER and spectral spreading.
A phase optimization scheme for proposed SI free PTS-OFDM system with
minimal computational complexity.
Conventional PTS scheme as well as proposed SI free PTS schemes requires
an exhaustive search for a set of phase factors to minimize the PAPR of
OFDM signal, which increases the computational complexity. Therefore, a
phase optimization scheme needs to be explored to reduce the number of
iterations to find optimal set of phase factors.
Generation of more number of candidate signals for M-2M mapping
based PTSOFDM systems to improve PAPR reduction capability.
M-2M mapping based PTS scheme uses two phase factors to limit the
constellation size extension, and provides same number of candidate
signals as conventional PTSOFDM system using two phase factors. The
PAPR performance of proposed M-

2M mapping based PTS scheme may be further improved at the cost of


increased computational complexity if the number of candidate signals is
increased. Therefore, there is a need to find the schemes to increase the
number of candidate signals in PTS based PAPR reduction using two phase
factors to further improve the PAPR reduction capability.
Analytical expressions for BER performance of OFDM system utilizing joint
ICI cancelation and PAPR reduction.
We have derived an analytical expression of PAPR performance of three
popular ICI cancellation schemes such as ICI self-cancellation scheme, new
ICI self-cancellation and ICI conjugate cancellation schemes. The analytical
expressions for BER performance for the proposed joint ICI cancellation and
PAPR reduction scheme still needs to find after exploring the feasibility of a
closed form expression for ICI power.

59 | P a g e
BIBLIOGRAPHY

1. "Digital Modulation Efficiencies". Barnard Microsystems. Archived from the original on 2011-04-30.
2. "Ciena tests 200G via 16-QAM with Japan-U.S. Cable Network". lightwave. April 17, 2014.
Retrieved 7 November 2016.
3. Kylia products Archived July 13, 2011, at the Wayback Machine., dwdm mux demux, 90 degree
optical hybrid, d(q) psk demodulatorssingle polarization
4. http://www.homeplug.org/media/filer_public/a1/46/a1464318-f5df-46c5-89dc-
7243d8ccfcee/homeplug_av2_whitepaper_150907.pdf Homeplug_AV2 whitepaper
5. http://www.itu.int/rec/T-REC-G.992.3-200904-I section 8.6.3 Constellation mapper - maximum number
of bits per constellation BIMAX 15
6. http://www.trangosys.com/products/point-to-point-wireless-backhaul/licensed-wireless/trangolink-
apex-orion.shtml A Apex Orion
7. http://www.trangosys.com/products/point-to-point-wireless-backhaul/licensed-wireless/trangolink-
apex-orion.shtml A Apex Orion
8. Cho, Kyongkuk; Yoon, Dongweon (July 2002). "On the General BER Expression of One- and Two-
Dimensional Amplitude Modulations". IEEE Trans. Commun. 50 (7): 1074
1080. doi:10.1109/TCOMM.2002.800818.
9. http://asp.eurasipjournals.com/content/2010/1/942638 DVB Hierarchical QAM constellation
10. Blahut, R. E. (1988). Principles and Practice of Information Theory. Boston, MA, USA: Addison
Wesley Publishing Company. ISBN 0-201-10709-0.
11. Howard Friedenberg and Sunil Naik. "Hitless Space Diversity STL Enables IP+Audio in Narrow STL
Bands" (PDF). 2005 National Association of Broadcasters Annual Convention. Retrieved April
17, 2005.
5. Jonqyin (Russell) Sun "Linear diversity analysis for QAM in Rician fading channels", IEEE WOCC
2014
The notation used here has mainly (but not exclusively) been taken from

John G. Proakis, "Digital Communications, 3rd Edition",

60 | P a g e