You are on page 1of 119



The purpose of a Communication System is to transport an information bearing signal

from a source to a user destination via a communication channel.


Fig. 1.1: Block diagram of Communication System

The three basic elements of every communication systems are

Receiver and
The Overall purpose of this system is to transfer information from one point (called Source)

to another point, the user destination.

The message produced by a source, normally, is not electrical. Hence an input transducer
is used for converting the message to a time varying electrical quantity called message
signal. Similarly, at the destination point, another transducer converts the electrical
waveform to the appropriate message.
The transmitter is located at one point in space, the receiver is located at some other point
separate from the transmitter, and the channel is the medium that provides the electrical
connection between them.
The purpose of the transmitter is to transform the message signal produced by the source
of information into a form suitable for transmission over the channel.
The received signal is normally corrupted version of the transmitted signal, which is due
to channel imperfections, noise and interference from other sources.The receiver has the
task of operating on the received signal so as to reconstruct a recognizable form of the
original message signal and to deliver it to the user destination.
Communication Systems are divided into 3 categories:
1. Analog Communication Systems are designed to transmit analog information using
analog modulation methods.
2. Digital Communication Systems are designed for transmitting digital information using
digital modulation schemes, and
3. Hybrid Systems that use digital modulation schemes for transmitting sampled and
quantized values of an analog message signal.


The figure 1.2 shows the functional elements of a digital communication system. Source of Information:1. An
2. Digital Information Sources.
Analog Information Sources Microphone actuated by a speech, TV Camera scanning a scene, contin
Digital Information Sources These are teletype or the numerical output of computer which consists of a se
An Analog information is transformed into a discrete information through the process of sampling and quant
Digital Communication System

Fig. 1.2: Block Diagram of Digital Communication System

The Source encoder ( or Source coder) converts the input i.e. symbol sequence
into a binary sequence of 0s and 1s by assigning code words to the symbols in the
input sequence. For eg. :-If a source set is having hundred symbols, then the
number of bits used to represent each symbol will be 7 because 2 7=128 unique
combinations are available. The important parameters of a source encoder are
block size,
code word lengths, average data rate and theefficiency of the coder (i.e. Actual
output data rate compared to the minimum achievable rate)
At the receiver, the source decoder converts the binary output of the channel
decoder into a symbol sequence. The decoder for a system using fixed length
code words is quite simple, but the decoder for a system using variable
words will be very complex.

Aimof thesource coding istoremovetheredundancy

minimized. Based on the probability of the symbol code word is assigned. Higher the
probability, shorter is the codeword. Ex: Huffman coding.
Error control is accomplished by the channel coding operation that consists of systematically addin
These extra bits do not convey any information but helps the receiver to detect and /

or correct some of the errors in the information bearing bits. There are two
methods of channel coding:
1. Block Coding: The encoder takes a block of k information bits from
the source encoder and adds r error control bits, where r is dependent on k and
error control capabilities desired.
2. Convolution Coding: The information bearing message stream is encoded in
a continuous fashion by continuously interleaving information bits and error
control bits.
The Channel decoder recovers the information bearing bits from the coded binary
stream. Error detection and possible correction is also performed by the channel
The important parameters of coder / decoder are: Method of coding, efficiency,
error control capabilities and complexity of the circuit.

The Modulator converts the input bit stream into an electrical waveform
suitable for transmission over the communication channel. Modulator can be
effectively used to minimize the effects of channel noise, tomatch the

spectrumoftransmittedsignalwithchannel characte
capability to multiplex many signals.

The extraction of the message from the information bearing waveform produced by the modulation
is bit stream. The important parameter is the method of demodulation.

The Channel provides the electrical connection between

The different channels are: Pair of wires, Coaxial cable, Optical fibre,
Radio channel, Satellite channel or combination of any of these.

The communication channels have only finite Bandwidth, non-ideal frequency

response, the signal often suffers amplitude and phase distortion as it travels over
the channel. Also, the signal power decreases due to the attenuation of the
channel. The signal iscorrupted by unwanted, unpredictable electrical signals
referred to as noise.
The important parameters of the channel are Signal to Noise power Ratio(SNR),
usable bandwidth, amplitude and phase response and the statistical properties of

Certain Issues in Digital Transmission

Sometimes, a comparison between two digital transmission systems is needed.
There are many parameters that can be used to compare between digital
transmission systems, but some of the most important parameters of a digital
transmission system are:

Rate (measured in bits per second): This is a measure of the
number of bits that can be transmitted over the communication channel per unit time.

Bandwidth Requirements
(measured in Hz): This is a measure of the
spectrum that the communication system requires to transmit the information at the desired transm

Error probability (measured

in percentages): This represents the percentage of
bits that are in error relative to the overall number of bits that are transmitted by the communicat

ransmission Power (or Bit Energy) (measured in Watts (or Jules/bit)): This
represents the amount of power of the transmitted signal that would be required to
achieve a particular desired error probability.

System Complexity (measured in cost of building or operating the system): This

represents that amount of money that a manufacturer will have to spend to build
the system and the amount of money that a user will have to pay to use the

The performance of a digital transmission system is a function of

the following factors:

1. Amount of energy in each digital bit (or pulse): Generally, the more energy a
digital bit (or pulse) has, the better the performance that the system will have.
2. The distance between the transmitter and receiver: Because energy is spread or
attenuated as it travels over the channel and more noise is added due to the
existence of more noise sources over long channels, generally the longer the path
that the digital transmitted signal has to travel, the worse the performance that the
system will have. However, you do not always have control over the distance
between the transmitted and receiver.
3. Amount of noise that is added to the signal: Certainly, the less the noise that is
added to the transmitted signal, the better the performance of the communication
system. We usually have limited control over the added noise.
4. Bandwidth of the transmission channel: By using larger bandwidth, we can

either transmit at a higher transmission bit rate while keeping the same probability o
bit error, or we can transmit at the same transmission bit rate but reduce the probab
communication system, the better the performance it will have.

Advantages of Digital Communication

The effect of distortion, noise and interference is less in a digital communication system. This is be
Regenerative repeaters can be used at fixed distance along the link, to identify and
regenerate a pulse before it is degraded to an ambiguous state.

3. Digital circuits are more reliable and cheaper compared to analog circuits.
4. The Hardware implementation is more flexible than analog hardware because of
the use of microprocessors, VLSI chips etc.
5. Signal processing functions like encryption, compression can be employed to
maintain the secrecy of the information.
6. Error detecting and Error correcting codes improve the system performance by
reducing the probability of error.
7. Combining digital signals using TDM is simpler than combining analog signals
using FDM. The different types of signals such as data, telephone, TV can be
treated as identical signals in transmission and switching in a digital
communication system.
8. We can avoid signal jamming using spread spectrum
technique. Disadvantages of Digital Communication:
1. Large System Bandwidth:- Digital transmission requires a large system
bandwidth to communicate the same information in a digital format as compared
to analog format.
2. System Synchronization:- Digital detection requires system synchronization
whereas the analog signals generally have no such requirement.

Channels for Digital Communications

The modulation and coding used in a digital communication system depend on the

characteristics of the channel.The two main characte

BANDWIDTH and POWER. In addition the other character

Five channels are considered in the digital communication, namely: telephone

channels, coaxial cables, optical fibers, microwave radio, and satellite channels.
Telephone channel: It is designed to provide voice grade communication. Also good for data comm
about 30db, and approximately linear response.

For the transmission of voice signals the channel provides flat amplitude
response. But for thetransmission of data and image transmissions, since the
phase delay variations are important an equalizer is used to maintain the flat
amplitude response and a linear phase response over the required frequency band.
Transmission rates upto16.8 kilobits per second have been achieved over the
telephone lines.
Coaxial Cable: The coaxial cable consists of a single wire conductor centered
inside an outer conductor, which is insulated from each other by a dielectric. The
main advantages of the coaxial cable are wide bandwidth and low
interference. But closely spaced repeaters are required. With repeaters spaced at
1km intervals the data rates of 274 megabits per second have been achieved.
Optical Fibers: An optical fiber consists of a very fine inner core made of silica
glass, surrounded by a concentric layer called cladding that is also made of glass.
The refractive index of the glass in the core is slightly higher than refractive index
of the glass in the cladding. Hence if a ray of light is launched into an optical fiber
at the right oblique acceptance angle, it is continually refracted into the core by
the cladding. That means the difference between the refractive indices of the core
and cladding helps guide the propagation of the ray of light inside the core of the
fiber from one end to the other.
Compared to coaxial cables, optical fibers are smaller in size and they offer
higher transmission bandwidths and longer repeater separations.
Microwave radio: A microwave radio, operating on the line-of-sight link, consists basically of a trans
30 GHz.

Under normal atm

reliable and prov

Satellite Channel: A Satellite channel consists of a satellite in geostationary orbit,

an uplink from ground station, and a down link to another ground station. Both
link operate at microwave frequencies, with uplink the uplink frequency higher
than the down link frequency. In general, Satellite can be viewed as repeater in
the sky. It permits communication over long distances at higher bandwidths and
relatively low cost.

Bandwidth is simply a measure of frequency range. The range of frequencies
contained in a composite signal is its bandwidth. The bandwidth is normally a
difference between two numbers. For example, if a composite signal contains
frequencies between 1000 and 5000, its bandwidth is 5000 - or 4000. If a range of
2.40 GHz to 2.48 GHz is used by a device, then the bandwidth would be 0.08
GHz (or more commonly stated as 80MHz).It is easy to see that the bandwidth we
define here is closely related to the amount of data you can transmit within it - the
more room in frequency space, the more data you can fit in at a given moment.
The term bandwidth is often used for something we should rather call a data rate,
as in my Internet connection has 1 Mbps of bandwidth, meaning it can
transmit data at 1

A message signal may originate from a digital or analog source. If the message signal is analog in n
Sampling operation is performed in accordance with the sampling theorem.

Sampling Theorem for Lowpass Signals

Part - I If a signal x(t) does not contain any frequency component beyond W Hz,
then the signal is completely described by its instantaneous uniform samples with
sampling interval (or period ) of Ts< 1/(2W) sec.

Part II The signal x(t) can be accurately reconstructed (recovered) from the set
of uniform instantaneous samples by passing the samples sequentially through an
ideal (brick-wall) lowpass filter with bandwidth B, where W B < fs W and fs
= 1/(Ts).
As the samples are generated at equal (same) interval (Ts) of time, the process of
sampling is called uniform sampling. Uniform sampling, as compared to any non-
uniform sampling, is more extensively used in time-invariant systems as the
theory of uniform sampling (either instantaneous or otherwise) is well developed
and the techniques are easier to implement in practical systems.

The concept of instantaneous sampling is more of a mathematical abstraction as

no practical sampling device can actually generate truly instantaneous samples (a
sampling pulse should have non-zero energy). However, this is not a deterrent in

using the theory of instantaneous sampling, as a fairly close approximation of

instantaneous sampling is sufficient for most practical systems. To contain our discussion on Nyquis

If x(t) represents a continuous-time signal, the equivalen

uniform samples {x(nTs)} may be represented as,

{x(nTs)} xs(t) = x(t).(t- nTs)(1)

where x(nTs) = x(t) t =nTs , (t) is a unit pulse singularity fun

Conceptually, one may think that the continuous-time signal x(t) is multiplied by
an (ideal) impulse train to obtain {x(nTs)} as in equation(1) can be rewritten as,

xs(t) = x(t). (t- nTs).

(2) Now, let X(f) denote the Fourier Transform F(T) of

x(t), i.e.
X ( f )= x (t ).exp( j 2 ft )dt . (3)

Now, from the theory of Fourier Transform, we know that the F.T of (t- nTs),
the impulse train in time domain, is an impulse train in frequency domain:

F{ (t- nTs)} = (1/Ts). (f- n/Ts) = fs. (f- nfs) (4)

If Xs(f) denotes the Fourier transform of the energy signal xs(t), we can write
using Eq. (1.2.4) and the convolution property:

Xs(f) = X(f)* F{ (t- nTs)}

= X(f)*[fs. (f- nfs)]

= fs.X(f)* (f- nfs)

= fs. nfs )d = fs. X().(f- nfs-)d = fs. X(f- nfs)
X(). (f .

[By sifting property of (t) and considering (f) as an even function, i.e. (f) = (-

This equation, when interpreted appropriately, gives an intuitive proof to Nyquists

theorems as stated above and also helps to appreciate their practical implications.
Let us note that while writing Eq.(5), we assumed that x(t) is an energy signal so
that its Fourier transform exists.
Further, the impulse train in time domain may be viewed as a periodic singularity
function with almost zero (but finite) energy such that its Fourier Transform [i.e. a
train of impulses in frequency domain] exists.
With this setting, if we assume that x(t) has no appreciable frequency component
greater than W Hz and if fs > 2W, then Eq.(1.2.5) implies that Xs(f), the Fourier
Transform of the sampled signal xs(t) consists of infinite number of replicas of
centered at discrete frequencies n.fs, - < n < and scaled by a constant fs= 1/Ts
(Fig. 1.3).

Fig. 1.3 Spectra of (a) an analog signal x(t) and (b

Fig. 1.3 indicates that the bandwidth of this instantaneously sampled wa

xs(t) is infinite while the spectrum of x(t) appears in a periodic manner, c

Now, Part I of the sampling theorem is about the condition fs > 2.W i.e. (fs W) >
W and ( fs + W) < W. As seen from Fig. 1.3, when this condition is satisfied, the spectra of xs(t),

of x(t) is present in xs(t) without any distortion. This implies that xs(t), the
appropriately sampled version of x(t), contains all information about x(t) and
thus represents x(t).

The second part of Nyquists theorem suggests a method of recovering x(t) from
its sampled version xs(t) by using an ideal lowpass filter. As indicated by dotted
lines in Fig. 1.3, an ideal lowpass filter (with brick-wall type response) with a
bandwidth W
B < (fs W), when fed with xs(t), will allow the portion of Xs(f), centered at f
= 0 and will reject all its replicas at f = n fs, for n 0. This implies that the shape
of the continuous-time signal xs(t), will be retained at the output of the ideal
Hartley Shannon Law

The theory behind designing and analyzing channel codes is called Shannons
noisy channel coding theorem. It puts an upper limit on the amount of information
you can send in a noisy channel using a perfect channel code. This is given by the
following equation:

where C is the upper bound on the capacity of the channel (bit/s), B is the bandwidth
of the channel (Hz) and SNR is the Signal-to-N ise ratio (unitless).

Bandwidth-S/N Tradeof
The expression of the channel capacity of the Gaussian channel makes intuitive sense:

As the bandwidth of the channel increases, it is possible to make faster

changes in the information signal, thereby increasing the information rate.
As S/N increases, one can increase the information rate while stillpreventing errors due to noise.

3. For no noise, S/N tends to infinity and an infinite information

rate is possibleirrespective of bandwidth.

Thus we may trade off bandwidth for SNR. For example, if S/N = 7 and B =
4kHz, then the channel capacity is C = 12 10 3 bits/s. If the SNR increases to S/N
= 15 and B is decreased to 3kHz, the channel capacity remains the same.
However, as B tends to 1, the channel capacity does not become infinite since,
with an increase in bandwidth, the noise power also increases. If the noise power
spectral density is /2, then the total noise power is N = B, so the Shannon-
Hartley law becomes
Pulse Code Modulation

In the simplest model of a telephone speech communication there is a direct, dedicated, physical c
In Pulse Amplitude Modulation (PAM), the unmodified electrical signal is not sent on to the connect

Because each sample is very short (~4s) there is a lot of time between samples
(~121s). Samples from other conversations are put into this spare time.
Usually the samples from 32 separate conversations are put on to a single line.
This process is called Time Division Multiplexing (TDM).

Each sample is very short, and will be distorted as it travels across a

communications network. In order to reconstruct the original analogue signal the
only information the receiver needs to have about a sample is its amplitude, but if
this is distorted then all information about the sample has been lost. To overcome
this problem, the pulse is not transmitted directly, instead its amplitude is
measured and converted into an 8 binary number - a sequence of 1s and 0s. At the
receiver end, the receiver merely needs to detect if a 1 or a 0 has been received so
that it can still recover the amplitude of a PAM pulse even if the 1s and 0s used to
describe it have been distorted.
The process of converting the amplitude of each pulse into a stream of 1s and 0s
is called Pulse Code Modulation (PCM)

Note that the process of PAM and PCM (but without the use of TDM) is
essentially used to store music and speech on CDs, but with a higher sample rate,
more bits per sample and complex error correction mechanisms.

Some terms are:

Sampling The process of measuring the amplitude of a continuous-time signal at

discrete instants. It converts a continuous-time signal to a discrete-
time signal.
Quantizing Representing the sampled values of the amplitude by a finite set of
levels. It converts a continuous-amplitude sample to a discrete-
amplitude sample.
Encoding Designating each quantized level by a (binary) code.
Sampling and quantizing operations transform an analogue signal to a digital signal.

Use of quantizing and encoding distinguishes PCM from analogue pulse

modulation methods.
The quantizing and encoding operations are usually performed in the same circuit
at the transmitter, which is called an Analogue to Digital Converter (ADC). At the
receiver end the decoding operation converts the (8 bit) binary representation of
the pulse back into an analogue voltage in a Digital to Analogue Converter (DAC)
Pulse Code Modulation
Pulse Code Modulation (PCM) is an extension of PAM wherein each analogue
sample value is quantized into a discrete value for representation as a digital code
Thus, as shown in Fig. 2.1 a PAM system can be converted into a PCM system by
adding a suitable analogue-to-digital (A/D) converter at the source and a digital-
to- analogue (D/A) converter at the destination.
M o d u la to r

A n a lo
gue P a ra lle l to S e rDiaigl ita l P u ls e
A to D B in a C o n v e r te r G e n e ra to r u tp u t
In p uS a m p le r C o n v e r te r
ry C o

D e m o d u la to r

P C M S e r ia l to P a ra lle l A n a lo g u e
In p uCt o n v e r te r D to A
C o n v e r te r O u tpt

PCM is a true digital process as compared to PAM. In PCM the speech signal is
converted from analogue to digital form.

PCM is standardised for telephony by the ITU-T (International Telecommunications Union - Telecoms
frequency response of the handset microphone has a sharp roll-off from 3.4 kHz.

In quantization the levels are assigned a binary codeword. All sample values
falling between two quantization levels are considered to be located at the centre
of the quantization interval. In this manner the quantization process introduces a
certain amount of error or distortion into the signal samples. This error known as
quantization noise, is minimised by establishing a large number of small
quantization intervals. Of course, as the number ofquantization intervals increase,
so must the number or bits increase to uniquely identify the quantization intervals.
For example, if an analogue voltage level is to be converted to a digital system
with 8 discrete levels or quantization steps three bits are required. In the ITU-T
version there are 256 quantization steps, 128 positive and 128 negative, requiring
8 bits. A positive level is represented by having bit 8 (MSB) at 0, and for a
negative level the MSB is 1.

The process of quantizing a signal is the first part of converting an sequence of

analog samples to a PCM code. In quantization, an analog sample with an
amplitude that may take value in a specific range is converted to a digital sample
with an amplitude that takes one of a specific predefined set of quantization
values. This is performed by dividing the range of possible values of the analog
samples into L different levels, and assigning the center value of each level to any
sample that falls in that quantization interval. The problem with this process is
that it approximates the value of an analog sample with the nearest of the
quantization values. So, for almost all samples, the quantized samples will differ
from the original samples by a small amount. This amount is called the
quantization error. To get some idea on the effect of this quantization error,
quantizing audio signals results in a hissing noise similar to what you would hear
when play a random signal.

Assume that a signal with power Psis to be quantized using a quantizer with L =
2n levels ranging in voltage from mp tomp as shown in the fig. 2.2

A quantization interval Corresponding quantization

value mp

0T Ts 3 Ts 4 Ts 5 Ts
s 2
L = 2n
L levels t
n bits

Q uantizer Input Sam ples x
Q uantizer O utput Sam ples x q

Fig. 2.2

We can define the variable v to be the height of the each of the L levels of the
quantizer as shown above. This gives a value of v equal to

Therefore, for a set of quantizers with the same mp, the larger the number of levels
of a quantizer, the smaller the size of each quantization interval, and for a set of
quantizers with the same number of quantization intervals, the larger mp is the
larger the quantization interval length to accommodate all the quantization range.

Now if we look at the input output characteristics of the quantizer, it will be

similar to the red line in the following figure. Note that as long as the input is
within the quantization range of the quantizer, the output of the quantizer
represented by the red line follows the input of the quantizer. When the input of
the quantizer exceeds the range of mp tomp, the output of the quantizer starts to
deviate from the input and the quantization error (difference between an input
and the corresponding output sample) increases significantly.

ut xq


Input x
v v v
v v v v v v





Fig. .2.3

Now let us define the quantization error represented by the difference between the
input sample and the corresponding output sample to be q, or

q x
x q.
Plotting this quantization error versus the input signal of a quantizer is seen next.
Notice that the plot of the quantization error is obtained by taking the difference
between the blow and red lines in the above Fig. 2.3

Quantization Error q

v/2 v v v Input x
v v v v v


Fig. 2.4

It is seen from the Fig 2.4 that the quantization error of any sample is restricted
between v/2 andv/2 except when the input signal exceeds the range of
quantization of mp to mp.

Uniform We assume that the amplitude of the signal m(t) is

Quantization confined to the range (-mp, +mp ). This range (2mp) is
divided into L levels, each of step size , given by
O u tp u t

= 2 mp / L

A sample amplitude value is approximated by the

-m p
midpoint of the interval in which it lies. The
+m p input/output characteristics of a uniform quantizer is
In p u t

shown in Fig. 2.5

Fig. 2.5

-Companding is the process of compressing and then expanding

-High amplitude analog signals are compressed prior to txn. and then expanded in
the receiver

-Higher amplitude analog signals are compressed and Dynamic range is improved

-Early PCM systems used analog companding, where as modern systems use
digital companding.

Fig 2.6 Basic companding process

Analog companding
2.7 PCM system with analog companding

--In the transmitter, the dynamic range of the analog signal is compressed, and
then converted o a linear PCM code.

--In the receiver, the PCM code is converted to a PAM signal, filtered, and then
expanded back to its original dynamic range.

-- There are two methods of analog companding currently being used that closely
approximate a logarithmic function and are often called log-PCM codes.

The two methods are 1) -law and

2) A-law

-law companding

ln 1 inV
ax V
ln 1

Fig. 2.8m-law companding

where Vmax = maximum uncompressed analog input amplitude(volts)

Vin = amplitude of the input signal at a particular instant of time (volts)

= parameter used tio define the amount of compression (unitless) Vout = compressed output am

A-law companding

--A-law is superior to -law in terms of small-signal quality

--The compression characteristic is given by

where y =
A | x| 1
, 0 | x | x = Vin
Vout y 1 log A /

Vmax 1 log( A | x |)
| x | 1
1 log A A
Digital Companding:

--With digital companding, the analog signal is first sampled and converted to a
linear PCM code, and then the linear code is digitally compressed.

-- In the receiver, the compressed PCM code is expanded and then decoded back
to analog.

-- The most recent digitally compressed PCM systems use a 12- bit linear PCM
code and an 8-bit compressed PCM code.

Digital compression error

--To calculate the percentage error introduced by digital compression



1.3 PCM Encoding Process (HDB3)

The output from the analogue to digital converter (ADC) has n parallel bits. In the case of telephony

the signed bit is 1. The remaining 7 bits are used to code the sample value. The
ITU- T define a look up table which allocates a particular binary code to each
quantified A-law value.

The line coding which is used assigns opposite polarities to successive 1s. This
eliminates any DC voltage on the line, and reduces the inter symbol interference
if adjacent bits are 1. If there is silence on the PCM channel then the measured
samples will be 0 Vrms and the output of the DAC will be 1000 0000. A stream of
all zeros is not desirable on an active channel because

all zeros could also be a fault condition and

it is difficult to recover the clock signal from the incoming signal.
The coding system HDB3 is used and was developed to eliminate all zeros, and to
assign opposite polarities to successive 1s.

This is a bipolar signalling technique (i.e. relies on the transmission of both

positive and negative pulses).

In AMI positive and negative pulses (of equal amplitude) are used for alternative
symbols 1. No pulse is used for symbol 0. In either case the pulse returns to 0
before the end of the bit interval. This eliminates any DC on the line.

HDB3 encoding rules follow those for AMI, except that a sequence of four
consecutive 0's are encoded using a special "violation" bit. The 4 th 0 bit is given
the same polarity as the last 1-bit which was sent using the AMI encoding rule.
This prevents long runs of 0's in the data stream which may otherwise prevent a
receiver from tracking the centre of each bit. By introducing violations, extra
"edges" are introduced, enabling a Digital PLL to reliably reconstruct the clock
signal at the receiver. The HDB3 is transparent to the sequence of bits being
transmitted (i.e. whatever data is sent, the Digital PLL can reconstruct the data
and extract the bits at the receiver).

To prevent a DC being introduced by excessive runs of zeros any run of more

than four zeros encodes as B00V. The value of B is assigned + or - alternately
throughout the bit stream.

Example 1111 1111 = + - + - + - + -


1010 1010 = + 0 - 0 + 0 - 0
B 0 B 0 B 0 B 0

1000 0001 + 0 0 0 + 0 0 -
= B 0 0 0 V 00B

1000 0110 = + 0 0 0 + - +0

= B 0 0 0 V BB0
PCM Timing and Synchronisation
The PCM receiver must be able to identify the start and finish of each full
sampling sequence and to identify each bit position. The sampling clock needs to
be either sent to, or regenerated at, the receiving side to determine when each full
sequence of sampling begins and ends. The data clock is also needed to determine
exactly when to read each bit of information.

d a ta c lo c k 6 4 k b i t / s

15.625 s

fra m e c lo c k 125
A PCM channel is sampled at 8,000 Hz or
once every 125 s. If there is one channel or
30 TDM channels the sampling period is
fixed at 125 s and this period is known as a
B1 B2
1 0
B3 B4 B5 B6 B7 B8 B1 Therefore the frame clock must have a
1 0 0 1 1 1 ?
period of 125 s. The rising edge of the
frame clock
Fig. 2.9 informs the receiver that the next bit will be
Bit 1 of a new sample. The falling edge of
data clock informs the receiver that it must read the data bit.

When the bit stream is transmitted along a line the pulses become distorted and
the rise and fall times become significant. Ideally, a 1 will be high for 15.625
s. In practice the pulse may only be above the high threshold for a few s so it
is very important that the bit is read within a certain time limit of the clock pulse.

The simplest way to synchronise a PCM sender to a PCM receiver is to send the
clock signals on different circuits to the data This would be done in a self-
contained system such as private branch exchange (PBX). Telephony is full
duplex so that there is a coder and a decoder at each port, but each would use the
same clock.

To minimise the number of circuits it is possible to use a line-coding scheme

which allows the receiver to extract the clocks from the PCM signal. In this case
the receiver will have free running clocks that lock (using a PLL) to the phase
and frequency of the transitions in the data stream. The line-coding scheme
ensures that there is a transition for every data bit.

Differential pulse coding schemes

PCM transmits the absolute value of the signal for each frame. Instead we can
transmit information about the difference between each sample. The two main
differential coding schemes are:
Delta Modulation
Differential PCM and Adaptive Differential PCM (ADPCM)
Delta Modulation
Delta modulation converts an analogue signal, normally voice, into a digital

g ra n u la r n

o ise 1 1 The analogue signal is sampled as in

1 the PCM process. Then the sample is
1 0 0 1

1 1 0
compared with the previous sample.
0 0
The result of the comparison is
0 0 0
quantified using a one bit coder. If the
sample is greater than the previous
sample a 1 is generated. Otherwise a 0
is generated.
The advantage of delta modulation
PCM is its simplicity and lower cost. But the noise performance is not as

Fig. 2.10
good as PCM.

To reconstruct the original from the quantization, if a 1 is received the signal is increased by a step
dx(t)/dt q /T = q * fs

where: x(t) = input signal, q = step size, T = period between samples, fs = sampling frequency

Assume that the input signal has maximum amplitude A and maximum frequency
F. The most rapidly changing input is provided by x(t) = A * sin (2 * * F * t).
For this dx(t)/dt = 2 * * F * A * sin (2 * * F *
t). This slope has a maximum value of 2 * * F * A
Overload occurs if 2 * * F * A> q
* fs To prevent overload we require q * fs> 2 * * F
Example A = 2 V, F = 3.4 kHz, and the signal is sampled 1,000,000 times
per second, requires q > 2 * 3.14 * 3,400 * 2 /1,000,000 V > 42.7 mV
Granular noise occurs if the slope changes more slowly than the step size. The
reconstructed signal oscillates by 1 step size in every sample. It can be reduced by
decreasing the step size. This requires that the sample rate be increased. Delta
Modulation requires a sampling rate much higher than twice the bandwidth. It
requires oversampling in order to obtain an accurate prediction of the next input,
since each encoded sample contains a relatively small amount of information.
Delta Modulation requires higher sampling rates than PCM.

Differential PCM (DPCM) and ADPCM

D iffe re n t ia Encod
A n a lo g to r Q u a n t is e d D iffe
B a n d L im itin
u e In erEnod re n c e S a
g F i l te r
put er m p le s


Acc um
u la to r D

Fig. 2.11

DPCM is also designed to take advantage of the redundancies in a typical speech

waveform. In DPCM the differences between samples are quantized with fewer
bits that would be used for quantizing an individual amplitude sample. The
sampling rate is often the same as for a comparable PCM system,

unlike Delta Modulation.

Adaptive Differential Pulse Code Modulation ADPCM is standardised by ITU-

T recommendations G.721 and G.726. The method uses 32,000 bits/s per voice
channel, as compared to standard PCMs 64,000 bits/s. Four bits are used to
describe each sample, which represents the difference between two adjacent
samples. Sampling is 8,000 times a second. It makes it possible to reduce the bit
flow by half while maintaining an acceptable quality. While the use of ADPCM
(rather than PCM) is imperceptible to humans, it can significantly reduce the
throughput of high speed modems and fax transmissions.

The principle of ADPCM is to use our knowledge of the signal in the past time to
predict the signal one sample period later, in the future. The predicted signal is
then compared with the actual signal. The difference between these is the signal
which is sent to line - it is the error in the prediction. However this is not
done by making
comparisons on the incoming audio signal - the comparisons are done after PCM

To implement ADPCM the original (audio) signal is sampled as for PCM to

produce a code word. This code word is manipulated to produce the predicted
code word for the next sample. The new predicted code word is compared with
the code word of the second sample. The result of this comparison is sent to line.
Therefore we need to perform PCM before ADPCM.

The ADPCM word represents the prediction error of the signal, and has no
significance itself. Instead the decoder must be able to predict the voltage of the
recovered signal from the previous samples received, and then determine the
actual value of the recovered signal from this prediction and the error signal, and
then to reconstruct the original waveform.

ADPCM is sometimes used by telecom operators to fit two speech channels onto a
single 64 kbit/s link. This was very common for transatlantic phone calls via satellite up until a few

6.9 Delta Modulation

Delta modulation, like DPCM is a predictive waveform coding technique and can
be considered as a special case of DPCM. It uses the simplest possible quantizer,
namely a two level (one bit) quantizer. The price paid for achieving the simplicity
of the quantizer is the increased sampling rate (much higher than the Nyquist rate)
and the possibility of slope-overload distortion in the waveform reconstruction, as
explained in greater detail later on in this section.

In DM, the analog signal is highly over-sampled in order to increase the adjacent
sample correlation. The implication of this is that there is very little change in two
adjacent samples, thereby enabling us to use a simple one bit quantizer, which like
in DPCM, acts on the difference (prediction error) signals.

In its original form, the DM coder approximates an input time function by a series
of linear segments of constant slope. Such a coder is therefore referred to as a
Linear (or non-adaptive) Delta Modulator (LDM). Subsequent developments have
resulted in delta modulators where the slope of the approximating function is a
variable. Such coders are generally classified under Adaptive Delta Modulation
(ADM) schemes. We use DM to indicate either of the linear or adaptive variety.

Fig. 2.12: Waveforms illustrative of LDM operation

Deltamodulation principleofoperation

atio n(PCM),whichrequiredadifficult-to-implementanalog-to-
moreforaspeechbandwidthof4 kHz)thevalueofeachbitbeing
epr evioussample.Itisan exampleofdifferentialpulsecodemodulation(DPCM).


eac omparisonofthecurrentsamplewiththatprecedingit,andtooutputasinglebit

which indicatesthesignofthe differencebetweenthe twosamples.Thisinprinciple

would requirea sample-and-hold type circuit.
ntb acktotheinputandthetwoanalogsignalscomparedinacomparator.Theoutput

isa delayedversionoftheinput,andsothecomparison
npri nciple.

ulat oryouwill bemodelling.
Thesystemisintheformofafeedbackloop.Thismeansthatitsoperationisnotn ecessaril
trivial.Butyoucanbuildit,andconfirmthatitdoesbehaveinthe manner
odi gitalconverter,andis thestarting pointfrom which
more sophisticateddeltamodulatorscanbe
iagrambeingeither Vvolts.Thisisthedeltamodulatedsignal,thewaveformof
whichisshowninFigure 2.Itisfedback,inafeedbackloop,viaanintegrator,toasummer.

2.15.Itis shownoverlaid uponthemessage,ofwhich itisanapproximation.

Figure 2.15:integrator output superimposed on the messagewith the delta modulated signal below

mer, andthe difference-anerror signal-isthe signalappearingatthe
eou tput.Themessagecomes in
the message,isadjustedwiththeamplifiertomatchitasclosely aspossible.

ons tantisanimportantparameter.
este p withinthesamplinginterval.

therectangularwavefromthesamplerisV volt.Forachangeofinputsampleto
positive,thechangeofintegrator output will be,

wherekisthegainoftheamplifier precedingthe integrator(asinFigure2.13).



amo dulatedsignal.
Theintegralofthebinarywaveformisthesawtoothapproximationto themessage.
wto othwave istheprimaryoutputfromthedemodulatoratthereceiver.
oxi mationprocessatthemodulator.
two kinds.Thesearedue toslopeoverload, and granularity.

-of- changeoftheinput signalinthe regionsofgreatestslope.
butt heapproximationispoorelsewhere.Thisisslopeoverload,duetotoosmalla
Slopeoverloadisillustrated inFigure2.16.
slo p e o v e rlo a d


mp ling rate).This isillustratedin
2.17.Thesawtoothisbetterabletomatchthemessageinthe regionsofsteep slope.

Figure2.17:increasedstepsize to reduce slope overload


2.18,wheretheratehasbeenincreasedbyafactorof2.4times, but thestep isthe same
size asinFigure2.15.
tim e

Figure2.18:increasedsampling rate to reduce slope overload

1.4 Granularnoise

ReferbacktoFigure 2.16.Thesawtoothfollowsthemessagebeingsampledquitewellintheregionsofsmalls


1.5 noiseanddistort

dby minimizingthequantizingerroratthesummer output, or

Adaptive Delata Modulation

The Operation Theory of ADM Modulation

From previous chapter, we know that the disadvantage of delta modulation is

when the input audio signal frequency exceeded the limitation of delta modulator,

Then this situation will produce the occurrence of slope overload and cause signal
distortion. However, the adaptive delta modulation (ADM) is the modification of
delta modulation to improve the disadvantage of the occurrence of slope overload.

Figure 2.20 is the block diagram of ADM modulator. In figure 2.20, we can see
that the delta modulator is comprised by comparator, sampler and integrator, then
the slope controller and the level detect algorithm comprise a quantization level
adjuster, which can control the gain of the integrator in the delta modulator. ADM
modulator is the modification of delta modulator, therefore, due to the delta
modulator has the problem of slope overload at low and high frequencies. The
reason is the magnitude of the (t) of delta modulator is fixed, i.e. the increment
of or - is unable to follow the variation of the slope of the input signal. When
the variation of the slope of the input signal is large, the magnitude of (t) still
can increase by following the variation, then this situation will not occur the
problem of slope overload. On the other hand, there is another technique, which is
known as continuous variable slope delta (CVSD) modulation. This technique is
commonly used in Bluetooth application. CVSD modulator is also the
modification of delta modulator, use to improve the occurrence of slope overload.
The different between the CVSD and ADM modulators are the quantization level
adjuster A. ADM modulator is discrete values and the quantization level adjuster
of CVSD modulator is continuous. Simply, the quantization value of ADM
modulator is the variation of digital, such as the quantization values of +1, +2, +3,
-2, -3 and so on. As for CVSD modulator, the quantization value is the variation
of analog, such as the quantization values of +1,
+1.1, +1.2, -1.5, -0.3, -0.9 and so on.
Fig. 2.20 The Operation Theory of ADM Modulation


Digital modulation techniques

Modulation is defined as the process by which some characteristics of a carrier
is varied in accordance with a modulating wave. In digital communications, the
modulating wave consists of binary data or an M-ary encoded version of it and
the carrier is sinusoidal wave.
Different Shift keying methods that are used in digital modulation techniques are

Amplitude shift keying [ASK]

Frequency shift keying [FSK]
Phase shift keying [PSK]
Fig 3.1 Different modulations
1. ASK[Amplitude Shift Keying]:

In a binary ASK system symbol 1 and 0 are transmitted as

S (t) 2E b
Cos2 f t for symbol 1
Tb 1

S 2 (t) 0 for symbol 0

2. FSK[Frequency Shift Keying]:

In a binary FSK system symbol 1 and 0 are transmitted as

S (t) 2Eb Cos2f t for symbol 1
1 1

S 2(t) 2Eb Cos2f 2t for symbol 0
3. PSK[Phase Shift Keying]:

In a binary PSK system the pair of signals S1(t) and S2(t) are
used to represent binary symbol 1 and 0 respectively.
S1 (t) 2Eb Cos2fc t --------- for Symbol 1

2Eb 2Eb
S2 (t) Cos(2fc t ) Cos2fc t ------- for Symbol 0
Tb Tb

Hierarchy of digital modulation technique

Digital Modulation Technique

Coherent Non - Coherent

Binary M - ary Hybrid Binary M - ary

(m) = 2 (m) = 2

* ASK M-ary ASK M-ary APK * ASK M-ary ASK

* FSK M-ary FSK M-ary QAM * FSK M-ary FSK
* PSK M-ary PSK * DPSK M-ary DPSK
Coherent Binary PSK:

Non Return to
Zero Level Product
Encoder Modulator
Binary Binary PSK Signal
Data Sequence

(t) 2 Cos2f t
1 c


Fig. 3.2 Block diagram of BPSK transmitter

x(t) dt x1 Decision Device Choose 1 if x1>0

Choose 0 if x1<0

1 (t) Threshold = 0

Fig 3.3 Coherent binary PSK receiver

In a Coherent binary PSK system the pair of signals S1(t) and S2(t) are used
to represent binary symbol 1 and 0 respectively.

S1 (t) Cos2fc t --------- for Symbol 1

2Eb 2Eb
S2 (t) Cos(2fc t ) Cos2fc t ------- for Symbol 0
Tb Tb
b0 b1
Where Eb= Average energy transmitted per bit Eb
In the case PSK, there is only one basic function of Unit energy which is given
(t) 2 Cos2f t 0tT
T c b

Therefore the transmitted signals are given by

S (t) E (t) 0 t T for 1
1 b 1 B

S (t) E (t) 0t for Symbol 0

2 b 1

A Coherent BPSK is characterized by having a signal space that

is one dimensional (N=1) with two message points (M=2)

S11 S1 (t)1 (t) dt


S21 S2 (t)1 (t) dt Eb

The message point corresponding to S1(t) is located at S11 Eb and S2(t) is located at S21 Eb
To generate a binary PSK signal we have to represent the input binary sequence in
polar form with symbol 1 and 0 represented by constant amplitude levels of

Eb &respectively.
b This signal transmission encoding is performed by a NRZ

level encoder.The resulting binary wave [in polar form] and a sinusoidal carrier 1 (t)
[whose frequency] fare
c applied to a product modulator. The desired BPSK wave

is obtained at the modulator output.

To detect the original binary sequence of 1s and 0s we apply the noisy PSK
signal x(t) to a Correlator, which is also supplied with a locally generated coherent
reference signal 1 (t) as shown in fig (b). The correlator output x1 is compared with a
threshold of zero volt.
If x1 > 0, the receiver decides in favour of
symbol 1. If x1 < 0, the receiver decides in
favour of symbol 0.
Coherent Binary FSK

In a binary FSK system symbol 1 and 0 are transmitted as

S (t) 2E Cos2f t for symbol 1

1 1
S 2(t) Cos2f 2t for symbol 0

The basic functions are given by

(t) 2 Cos2ft And

1 T 1

(t) 2 Cos2ft for 0 t T And Zero Otherwise

2 2 b

Therefore FSK is characterized by two dimensional signal space with two

message points i.e. N=2 and m=2.

The two message points are defined by the signal vector

Generation and Detection:-

Fig. 3.4: (a) FSK transmitter (

A binary FSK Transmitter is as shown in fig. (a). The incoming binary data
sequence is applied to on-off level encoder. The output of encoder is Eb volts for symbol 1 and 0 vo
switched on with oscillator frequency f1, for symbol 0, because of inverter the lower
channel is switched on with oscillator frequency f2. These two frequencies are combined using an a
The detector consists of two correlators. The incoming noisy BFSK signal x(t) is
common to both correlator. The Coherent reference signal 1 (t) and 2 (t) are supplied

to upper and lower correlators respectively.

The correlator outputs are then subtracted one from the other and resulting
random vector l (l=x1 - x2). The output l is compared with threshold of zero
If l > 0, the receiver decides in favour of
symbol 1. l < 0, the receiver decides in
favour of symbol0

Binary Binary ASK

Data ON-OFF Product Modulator Signal
Sequence Level
(t) Tb Cos2f t
1 e

Fig. 3.5 BASK transmitter


dt Decision Device
x(t) X 0 If x > choose symbol 1

If x < choose symb

1 (t)Thresho
Fig. 3.6 Coherent binary ASK demodulator In Coherent binary ASK system the basic function is give

(t) 2 Cos2f t0 t T
1 e b

The transmitted signals S1(t) and S2(t) are given by
S1 (t) Eb 1 (t) for Symbol 1

S2 (t) 0 for Symbol 0

The BASK system has one dimensional signal space with two messages (N=1,
Region E2 Region E1

Point 2
1 (t)
0 b
2 Point 1
Fig. 3.7 Signal Space representation of BASK signal

In transmitter the binary data sequence is given to an on-off encoder. Which gives
output Eb volts for symbol 1 and 0 volt for symbol 0. The resulting binary wave [in unipolar
form] and sinusoidal carrier 1 (t) are applied to a product modulator. The desired BASK wave is ob

In demodulator, the received noisy BASK signal x(t) is apply to correlator with coher
reference signal 1 (t) as shown in fig. (b). The correlator output x is compared with
If x > the receiver decides in favour of symbol 1. If x < the receiver decides in fa

Incoherent detection:

Fig. 3.8 : Envelope detector for OOK BASK

Incoherent detection as used in analog communication does not require carrier for
reconstruction. The simplest form of incoherent detector is the envelope detector as shown in
Fig. 3.8. The output of envelope detector is the baseband signal. Once the baseband signal is
recovered, its samples are taken at regular intervals and compared with threshold.
If Z(t) is greater than threshold ( ) a decision will be made in favour of symbol 1

If Z(t) the sampled value is less than threshold ( ) a decision will be made in favour
symbol 0.

Non- Coherenent FSK Demodulation:-

Fig. 3.9 : Incoherent detection of FSK

Fig. 3.9 shows the block diagram of incoherent type FSK demodulator. The detector
consists of two band pass filters one tuned to each of the two frequencies used to
communicate 0s and 1s., The output of filter is envelope detected and then baseband detected u

sinusoids is stronger at the receiver. If we take the difference of the outputs of the two envelope
detectors the result is bipolar baseband.

The resulting envelope detector outputs are sampled at t = kTb and their values are
compared with the threshold and a decision will be made infavour of symbol 1 or 0.

Differential Phase Shift Keying:- [DPSK]

(a) DPSK Transmitter

(b) DPSK Receiver Fig. 3.10 DPSK

A DPSK system may be viewed as the non coherent version of the PSK. It eliminates the need for co
Differential encoding of the input binary wave and
Phase shift keying

Hence the name differential phase shift keying [DPSK]. To send symbol 0
we phase advance the current signal waveform by 180 and to send symbol 1 we
leave the phase of the current signal waveform unchanged.
The differential encoding process at the transmitter input starts with an
arbitrary first but, securing as reference and thereafter the differentially encoded
sequence{dk} is generated by using the logical equation.

d k d k 1 bk d k 1 bk

Where bk is the input binary digit at time kTb and dk-1 is the previous value of
differentially encoded digit. Table illustrate the logical operation involved in the
generation of DPSK signal.

Input Binary Sequence {bK} 1 0 0 1 0 0 1 1

Differentially Encoded 1 1 0 1 1 0 1 1 1
sequence {dK}
Transmitted Phase 0 0 0 0 0 0 0
Received Sequence 1 0 0 1 0 0 1 1
(Demodulated Sequence)

A DPSK demodulator is as shown in fig(b). The received signal is first passed through a BPF centere
to the cosine of the difference between the carrier phase angles in the two correlator
inputs. The correlator output is finally compared with threshold of 0 volts . If correlator output is +

Fig. 3.11(a) QPSK Transmitter

Fig. 3.11(b) QPSK Receiver

In case of QPSK the carrier is given by

s (t) 2E Cos[2f t (2i 1) / 4] 0 t T i 1 to 4

T c

s (t) 2E Cos[(2i 1) / 4]cos( 2f t) 2E sin[(2i 1) / 4]sin( 2f t) 0 t T i 1 to 4

T c
T c
Fig. 3.11(c) QPSK Waveform

In QPSK system the information carried by the transmitted signal is contained

in the phase. The transmitted signals are given by
Where the carrier frequency f C
for some fixed integer nc

E = the transmitted signal energy per symbol.

T = Symbol duration.

The basic functions 1 (t) and 2 (t) are given by

1 (t) T cos 2 fc
0 t T


2(t) T b sin 2 f c 0 t T
There are four message points and the associated signal vectors are defined by

E cos
2i 1

4 i 1,2,3,4

E sin 2i
The table shows the elements of signal vectors, namely Si1 & Si2

Input dibit Phase of Coordinates of message
QPSK points
signal(radians) Si1 Si2

10 E E
4 2 2

00 3 E E
4 2 2

01 5 E E
4 2 2

11 7 E E
4 2 2

Therefore a QPSK signal is characterized by having a two dimensional signal constellation(i.e.N=2)a

.Fig 3.11(d) Signal-space diagram of coherent QPSK system.

Unit III

Base Band Transmission and Optimal Reception of Digital Signal

Pulse Shaping for Optimum Transmissions
Base Band Reception Techniques

Receiving Filter:
Correlative receiver
For an AWGN channel and for the case when the transmitted signals are
equally likely, the optimum receiver consists of two subsystems

1) Receiver consists of a bank of M product-integrator or

correlators 1(t) ,2(t) .M(t) orthonormal function
2) The bank of correlator operate on the received signal x(t) to produce
observation vector x

Implemented in the form of maximum likelihood detector that operates

on observationvector x to produce an estimate of the transmitted symbol m ii = 1
to M, in a way that would minimize the average probability of symbol error.
The N elements of the observation vector x are first multiplied by the
corresponding N elements of each of the M signal vectors s1, s2 sM , and
the resulting products are successively summed in accumulator to form the
corresponding set of
Inner products {(x, sk)} k= 1, 2 ..M. The inner products are corrected for the
fact that the transmitted signal energies may be unequal. Finally, the largest in
the resulting set of numbers is selected and a corresponding decision on the
transmitted message made.
The optimum receiver is commonly referred as a correlation receiver

Science each of the orthonormal basic functions are 1(t) ,2(t) .M(t) is
assumed to be zero outside the interval 0<t<T. we can design a linear filter with
impulse response hj(t), with the received signal x(t) the fitter output is given by
the convolution integral

yj(t) = xj

Where xj is the j th correlator output produced by the received signal x(t).

A filter whose impulse response is time-reversed and delayed version of the

input signal is said to be matched to xj (t)correspondingly, the optimum
receiver based on this isreferred as the matched filter receiver.

For a matched filter operating in real time to be physically realizable, it must be

(t) = input signal

h(t) = impulse response

W(t) =white noise

The impulse response of the matc

version of the input signal



The spectrum of the output signal of a matched filter with the matched signal as inpu


The output signal of a Matched Filter is proportional to a shifted version of the

autocorrelation function of the input signal to which the filter is matched.


The output Signal to Noise Ratio of a Matched filter depends only on the ratio
of the signal energy to the power spectral density of the white noise at the filter

The Matched Filtering operation may be separated into two matching
conditions; namely spectral phase matching that produces the desired output
peak at time T, and the spectral amplitude matching that gives this peak value its
optimum signal to noise density ratio.


The quality of digital transmission systems are evaluated using the bit
error rate. Degradation of quality occurs in each process modulation,
transmission, and detection. The eye pattern is experimental method that
contains all the information concerning the degradation of quality. Therefore,
careful analysis of the eye pattern is important in analyzing the degradation

Eye patterns can be observed using an oscilloscope. The received wave is

applied to the vertical deflection plates of an oscilloscope and the
sawtooth wave at a rate equal to transmitted symbol rate is applied to the
horizontal deflection plates, resulting display is eye pattern as it
resembles human eye.

The interior region of eye pattern is called eye opening

We get superposition of successive symbol intervals to produce eye pattern as

shown below.
The width of the eye opening defines the time interval over which the
received wave can be sampled without error from ISI

The optimum sampling time corresponds to the maximum eye opening

The height of the eye opening at a specified sampling time is a measure

of the margin over channel noise.

The sensitivity of the system to timing error is determined by the rate of closure
of the eye as the sampling time is varied.

Any non linear transmission distortion would reveal itself in an asymmetric or

squinted eye. When the effected of ISI is excessive, traces from the upper
portion of the eye pattern cross traces from lower portion with the result that the
eye is completely closed.

Information and Entropy

Although it is in principle a very old concept, entropy is generally credited to Shannon
because it is the fundamental measure in information theory. Entropy is often defined as an

where 0 log(0) = 0. The base of the logarithm is generally 2. When this is the case, the units
of entropy are bits.

Entropy captures the amount of randomness or uncertainty in a variable. This, in turn,is

aeasure of the average length of a message that would have to be sent to describe a sample.
Recall our fair coin from 1. Its entropy is:0.5log0.5 + 0.5log0.5= 1; that is, thereis one
bit of information in the random variable.This means that on average we need to sendone bit per tr
Even if I flip this coin 100 times, it doesnt matter because the outcome is always heads. I dont need to

There are other possibilities besides being completely random and completely deter-mined.
Imagine a weighted coin, such that heads occurr 75% of the time. The entropy would be:
0.75log0.75 + 0.25log0.25= 0.8113. After 100 trials, Id only need a message of about 82 bits
on average to describe the sample. Shannon showed that there exists a coder that can construct
messages of length H(X)+1, nearly matching this ideal rate.

Just as with probabilities, we can compute joint and conditional entropies. Joint
entropy is the randomness contained in two variables, while conditional entropy is a measure

of the randomness of one variable given knowledge of another. Joint entropy is defined as:
while the conditional entropy is:

There are several interesting facts that follow from these definitions. For example, two
random variables, X and Y, are considered independent if and only if HY| X=

or HXY= HX+HY It is also the case that HY|XHY (knowing more information
can never increase our uncertainty). Similarly, HXYHX+HY It is alsothe case that
HXY=HY|X+HX=HX Y+HY These relations hold in thegeneral case of more than
two variables.

There are several facts about discrete entropy, H(), that do not hold for continuous
ordifferential entropy, h(). The most important is that while H X 0 h() can actually be negative.
of randomness, it is still that case that if h X h Y then X has more randomness than Y.


Although co
not an adeq

us a great deal about Y or that H(Y) is small to begin with. Thus, we measure dependence
using mutual information:


Mutual information is a measure of the reduction of randomness of a variable given

knowledge of another variable. Using properties of logarithms, we can derive several equiva-
lent definitions:
= HXHX | Y

In addition to the definitions above, it is useful to realize that mutual information is a

particular case of the Kullback-Leibler divergence. The KL divergence is defined as:

KL divergence measures the difference between two distributions. It is sometimes called the
relative entropy. It is always non-negative and zero only when p=q; however, it is not a
distance because it is not symmetic.

In terms of KL divergence, mutual information is:

In other words, mutual information is a measure of the difference between the joint
probability and product of the individual probabilities. These two distributions are equivalent

only when X and Y are independent, and diverge as X and Y become more dependent.

Shannon-Fano Code
ShannonFano coding, named after Claude Elwood Shannon and Robert Fano, is a technique
for constructing a prefix code based on a set of symbols and their probabilities. It is
suboptimal in the sense that it does not achieve the lowest possible expected codeword length
like Huffman coding; however unlike Huffman coding, it does guarantee that all codeword
lengths are within one bit of their theoretical ideal I(x) =log P(x).
In ShannonFano coding, the symbols are arranged in order from most probable to least
probable, and then divided into two sets whose total probabilities are as close as possible to
being equal. All symbols then have the first digits of their codes assigned; symbols in the first
set receive "0" and symbols in the second set receive "1". As long as any sets with more than
one member remain, the same process is repeated on those sets, to determine successive
digits of their codes. When a set has been reduced to one symbol, of course, this means the
symbol's code is complete and will not form the prefix of any other symbol's code.

The algorithm works, and it produces fairly efficient variable-length encodings; when the two
smaller sets produced by a partitioning are in fact of equal probability, the one bit of
information used to distinguish them is used most efficiently. Unfortunately, ShannonFano
does not always produce optimal prefix codes.

For this reason, ShannonFano is almost never used; Huffman coding is almost as
computationally simple and produces prefix codes that always achieve the lowest expected code w
performance and minimum requirements for programming.


A Shannon
code table

counts so that each symbols relative frequency of occurrence is known.

Sort the lists of symbols according to frequency, with the most frequently occurring
symbols at the left and the least common at the right.

Divide the list into two parts, with the total frequency counts of the left part being as
close to the total of the right as possible.
The left part of the list is assigned the binary digit 0, and the right part is assigned the
digit 1. This means that the codes for the symbols in the first part will all start with 0,
and the codes in the second part will all start with 1.
Recursively apply the steps 3 and 4 to each of the two halves, subdividing groups and
adding bits to the codes until each symbol has become a corresponding code leaf on the tree.


The source of information A generates the symbols {A0, A1, A2, A3 and A4} with the
corresponding probabilities {0.4, 0.3, 0.15, 0.1 and 0.05}. Encoding the source symbols using
binary encoder and Shannon-Fano encoder gives:

Source Symbol Pi Binary Code Shannon-Fano

A0 0.4 000 0
A1 0.3 001 10
A2 0.15 010 110
A3 0.1 011 1110
A4 0.05 100 1111
Lavg H = 2.0087 3 2.05
Shannon-Fano code is a top-down approach. Constructing the code tree, we get
Source Coding

All source models in information theory may be viewed as random process or

random sequence models. Let us consider the example of a discrete memory less
source (DMS), which is a simple random sequence model.

A DMS is a source whose output is a sequence of letters such that each letter is

independently selected from a fixed alphabet consisting of letters; say a 1, a2 ,

.ak. The letters in the source output sequence are assumed to be random and

independent of each other. A fixed probability assignment for the occurrence of
letter is also assumed. Let us, consider a small example to appreciate the importance of
probability assignment of the source letters.

Let us consider a source with four letters a1, a2andwith

, aa34 P(a1)=0.5,

P(a2)=0.25, P(a3)= 0.13, P(a4)=0.12. Let us decide to go for binary coding of these four
source letters. While this can be done in multiple ways, two encoded representations are shown be

Code Representation#1: a1: 00, a2:01, a3:10, a4:11

Code Representation#2: a1: 0, a2:10, a3:001, a4:110

It is easy to see that in method #1 the probability assignment of a source letter

has not been considered and all letters have been represented by two bits each.
However in

the second method only a1 has been encoded in one bit, a2 in two bits and the
remaining two in three bits. It is easy to see that the average number of bits to be used
per source letter for the two methods are not the same. ( a for method #1=2 bits per
letter and a for
method #2 < 2 bits per letter). So, if we consider the issue of encoding a long sequence of
letters we have to transmit less number of bits following the second method. This is
an important aspect of source coding operation in general. At this point, let us note the
a) We observe that assignment of small number of bits to more probable letters and
assignment of larger number of bits to less probable letters (or symbols) may lead to
efficient source encoding scheme.
b) However, one has to take additional care while transmitting the encoded letters. A
careful inspection of the binary representation of the symbols in method #2 reveals
that it may lead to confusion (at the decoder end) in deciding the end of binary
representation of a letter and beginning of the subsequent letter.

So a source-encoding scheme should ensure that

The average number of coded bits (or letters in general) required per source letter is as
small as possible and
The source letters can be fully retrieved from a received encodedsequence.
In the following we discuss a popular variable-length source-coding scheme satisfying the above tw

Variable length Coding

Let us assume that a DMS U has a K- letter alphabet

{a1,a2, .aK} with

probabilities P(a 1), P(a2),. P(aK). Each source letter is to be encoded into a
codeword made of elements (or letters) drawn from a code alphabet containing D
symbols. Often for ease of implementation a binary code alphabet (D = 2) is chosen.
we observed earlier in an example, different codeword may not have same number of
code symbols. If nk denotes the number of code symbols corresponding to the source

letter ak , the average number of code letters per source letter ( n ) is:

n=P (ak)nk 5.1

k =1
Intuitively, if we encode a very long sequence of letters from a DMS, the
number of code letters per source letter will be close to n .

Now, a code is said to be uniquely decodable if for each source sequence of

finite length, the sequence of code letters corresponding to that source sequence is
different from the sequence of code letters corresponding to any other possible source

We will briefly discuss about a subclass of uniquely decodable codes, known

prefix condition code. Let the code word in a code be represented as

xk=(xk,1,xk,2,......,xk,nk), wherexk,1,xk,2,......,xk,nkdenote the individualcode letters

(when D=2, these are 1 or 0). Any sequence made up of an initial part of x k
that is xk,1,xk,2,......,xk,i for i nk is called a prefix of xk .

A prefix condition code is a code in which no code word is the prefix of

any other codeword.

Example: consider the following table and find out which code out of the four shown is
/are prefix condition code. Also determine n for each code.

a1, a2, a3 and

Source letters:- a4
P(ak) :- P(a1)=0.5, P(a2)=0.25, P(a3)= 0.125, P(a4)=0.125

Code Representation#1: a1: 00, a2:01, a3:10, a4:11

Code Representation#2: a1: 0, a2:1, a3:00, a4:11

Code Representation#3: a1: 0, a2:10, a3:110, a4:111

Code Representation#4: a1: 0, a2:01, a3:011, a4:0111

A prefix condition code can be decoded easily and uniquely. Start at the beginning of
sequence and decode one word at a time. Finding the end of a code word is not a
problem as the present code word is not a prefix to any other codeword.

Example: Consider a coded sequence 0111100 as per Code Representation #3 of

theprevious example. See that the corresponding source letter sequence is a1 ,a4 , a2 , a1 .

Now, we state one important theorem known as Kraft Inequality theorem


Binary Huffman Coding (an optimum variable-length source coding scheme)

In Binary Huffman Coding each source letter is converted into a binary code
word. It is a prefix condition code ensuring minimum average length per source letter in bits.

Let the source letters a1, a 2, .aK have probabilities P(a1), P(a2),.

P(aK) and let us assume that P(a1) P(a2) P(a 3). P(aK).

We now consider a simple example to illustrate the steps for Huffman coding.

Steps to calculate Huffman Coding

Example Let us consider a discrete memoryless source with six letters having

P(a1)=0.3,P(a2)=0.2, P(a 3)=0.15, P(a 4)=0.15, P(a5)=0.12 and P(a6)=0.08.

Arrange the letters in descending order of their probability (here they are
Consider the last two probabilities. Tie up the last two probabilities. Assign,

say, 0 to the last digit of representation for the least probable letter (a6) and

1 to the
last digit of representation for the second least probable letter (a5). That is,

assign 1 to the upper arm of the tree and 0 to the

lower arm.


2 0.2

P(a6)=0.08 0

(3) Now, add the two probabilities and imagine a new letter, say b1, substituting

a6 and a5. So P(b1) =0.2. Check whether a4 and b1are the least likely letters. If

not, reorder the letters as per Step#1 and add the probabilities of two least
For our example, it leads to:
P(a1)=0.3, P(a2)=0.2, P(b1)=0.2, P(a3)=0.15 and P(a4)=0.15

(4) Now go to Step#2 and start with the reduced ensemble consisting of a1 , a2 , a3 ,

a4 and b1. Our example results in:

P(a3)=0.15 1

Here we imagine another letter b1, with P(b2)=0.3.

t Continue till the first digits of the most reduced ensemble of two letters are
assigned a 1 and a 0.

Again go back to the step (2): P(a1)=0.3, P(b2)=0.3, P(a2)=0.2 and

Now we consider the last two probabilities:



So, P(b3)=0.4. Following Step#2 again, we get, P(b3)=0.4, P(a1)=0.3 and


Next two probabilities lead

to: 1



P(b2)=0.3 0

with P(b4) = 0.6. Finally we get only two probabilities

P(b4)=0.6 1


6. Now, read the code tree inward, starting from the root, and construct the
codewords. The first digit of a codeword appears first while reading the code
tree inward.
Hence, the final representation is: a1=11, a2=01, a3=101, a4=100, a5=001, a6=000.

A few observations on the preceding example

1. The event with maximum probability has least number of bits.

2. Prefix condition is satisfied. No representation of one letter is prefix for other.

Prefix condition says that representation of any letter should not be a part of
any other letter.
3. Average length/letter (in bits) after coding is

= P (ai )ni = 2.5 bits/letter.


4. Note that the entropy of the source is: H(X)=2.465 bits/symbol. Average length
per source letter after Huffman coding is a little bit more but close to the source
entropy. In fact, the following celebrated theorem due to C. E. Shannon sets the limiting value of av


Convolutional codes are commonly described using two parameters: the code
rate and the constraint length. The code rate, k/n, is expressed as a ratio of the number
of bits into the convolutional encoder (k) to the number of channel symbols output by
the convolutional encoder (n) in a given encoder cycle. The constraint length
parameter, K, denotes the "length" of the convolutional encoder, i.e. how many k-bit
stages are available to feed the combinatorial logic that produces the output symbols.
Closely related to K is the parameter m, which indicates how many encoder cycles an
input bit is retained and used for encoding after it first appears at the input to the
convolutional encoder. The m parameter can be thought of as the memory length of
the encoder.

Convolutional codes are widely used as channel codes in practical

communication systems for error correction. The encoded bits depend on the current k
input bits and a few past input bits. The main decoding strategy for convolutional
codes is based on the widely used Viterbi algorithm. As a result of the wide
acceptance of convolutional codes, there have been several approaches to modify and
extend this basic coding scheme. Trellis coded modulation (TCM) and turbo codes are
two such examples. In TCM, redundancy is added by combining coding and
modulation into a single operation. This is achieved without any reduction in data rate
or expansion in bandwidth as required by only error correcting coding schemes.
A simple convolutional encoder is shown in Fig. 7.1. The information bits are
fed in small groups of k-bits at a time to a shift register. The output encoded bits are
obtained by modulo-2 addition (EXCLUSIVE-OR operation) of the input information
bits and the contents of the shift registers which are a few previous information bits.

If the encoder generates a group of n encoded bits per group of k information

bits, the code rate R is commonly defined as R = k/n. In Fig. 7.1, k = 1 and n = 2.
The number, K of elements in the shift register which decides for how many
codewords one information bit will affect the encoder output, is known as the
constraint length of the code. For the present example, K = 3.

The shift register of the encoder is initialized to all-zero-state before encoding

operation starts. It is easy to verify that encoded sequence is 00 11 10 00 01
.for an input message sequence of 01011.
Fig. 7.1A convolutional encoder with k=1, n=2 and r=1/2

The operation of a convolutional encoder can be explained in several but

equivalent ways such as, by a) state diagram representation, b) tree diagram
representation and c) trellis diagram representation.

a) State Diagram Representation

A convolutional encoder may be defined as a finite state machine. Contents of
the rightmost (K-1) shift register stages define the states of the encoder. So, the
encoder in Fig. 7.2has four states. The transition of an encoder from one state to
another, ascaused by input bits, is depicted in the state diagram. Fig. 7.2 shows the
state diagram of the encoder in Fig. 7.1. A new input bit causes a transition from
one state to another. The
path information between the states, denoted as b/c1c2 , represents input information bit
b and the corresponding output bits (c1c2). Again, it is not difficult to verify from the
state diagram that an input information sequence b = (1011) generates an encoded
sequence c = (11, 10, 00, 01).

Fig. 7.2 State diagram representation for the encoder in Fig. 7.1
b) Tree Diagram Representation
The tree diagram representation shows all possible information and encoded
sequences for the convolutional encoder. Fig. 7.3 shows the tree diagram for the
encoder in Fig. 7.1. The encoded bits are labeled on the branches of the tree. Given an
input sequence, the encoded sequence can be directly read from the tree. As an
example, an input sequence (1011) results in the encoded sequence (11, 10, 00, 01).
c) Trellis Diagram Representation
The trellis diagram of a convolutional code is obtained from its state diagram.
All state transitions at each time step are explicitly shown in the diagram to retain the
time dimension, as is present in the corresponding tree diagram. Usually, supporting
descriptions on state transitions, corresponding input and output bits etc. are labeled in
the trellis diagram. It is interesting to note that the trellis diagram, which describes the
operation of the encoder, is very convenient for describing the behavior of
thecorresponding decoder, especially when the famous Viterbi Algorithm (VA) is
followed. Figure 7.4 shows the trellis diagram for the encoder in Figure 7.1.

Fig. 7.3 A tree diagram for the encoder in Fig. 7.1

Fig. 7.4Trellis diagram, used in the decoder corresponding to the encoder in
Hard-Decision and Soft-Decision Decoding
Hard-decision and soft-decision decoding are based on the type of
quantization used on the received bits. Hard-decision decoding uses 1-bit quantization
on the received samples. Soft-decision decoding uses multi-bit quantization (e.g. 3
bits/sample) on the received sample values.

Hard-Decision Viterbi Algorithm

The Viterbi Algorithm (VA) finds a maximum likelihood (ML) estimate of a
transmitted code sequence c from the corresponding received sequence r by
maximizing the probability p(r|c) that sequence r is received conditioned on the
estimated code sequence c. Sequence c must be a valid coded sequence.
The Viterbi algorithm utilizes the trellis diagram to compute the path metrics. The channel is assum
metric of all branches, associated with all the states are calculated similarly Now, at each depth of

process, the accumulated path metric is updated by adding the metric of the
incoming branch with the accumulated path metric of the state from where the
branch originated. No decision about a received codeword is taken from such
operations and the decoding decision is deliberately delayed to reduce the possibility
of erroneous decision.

The basic operations which are carried out as per the hard-decision Viterbi
Algorithm after receiving one codeword are summarized below:
All the branch metrics of all the states are determined;

Accumulated metrics of all the paths (two in our example code) leading to a
state are calculated taking into consideration the accumulated path metrics of
the states from where the most recent branches emerged;

Only one of the paths, entering into a state, which has minimum accumulated
path metric is chosen as the survivor path for the state (or, equivalently

So, at the end of this process, each state has one survivor path. The history
of a survivor path is also maintained by the node appropriately ( e.g. by storing
the codewords or the information bits which are associated with the branches
making the path);

(5) Steps a) to d) are repeated and decoding decision is delayed till sufficient number
of codewords has been received. Typically, the delay in decision making = Lx k codewords where L

The above procedure is repeated for each received codeword hereafter. Thus, the decision for a cod

Soft-Decision Viterbi Algorithm

In soft-decision decoding, the demodulator does not assign a 0 or a
1 to each received bit but uses multi-bit quantized values. The soft-decision Viterbi
algorithm is very similar to its hard-decision algorithm except that squared Euclidean
distance is used in the branch metrics instead of simpler Hamming distance. However,
the performance of a soft-decision VA is much more impressive compared to its HDD
(Hard Decision Decoding) counterpart [Fig. 7.6 (a) and (b)]. The computational
requirement of a Viterbi decoder grows exponentially as a function of the constraint
length and hence it is usually limited in practice to constraint lengths of K = 9.
Fig. 7.6 (a) Decoded BER vs input BER for the rate half convolutional
codes withViterbi Algorithm ; 1) k = 3 (HDD),2) k = 5 (HDD),3) k = 3
(SDD), and 4) k= 5 (SDD). HDD: Hard Decision Decoding; SDD: Soft
Decision Decoding.
Fig. 7.6 (b) Decoded BER vs Eb/No(in dB) for the rate half convolutional codes withViterbi
Algorithm ; 1) Uncoded system; 2) with k = 3 (HDD) and 3) k = 3 (SDD). HDD: Hard
Decision Decoding; SDD: Soft Decision Decoding.

Spread Spectrum Modulation


Initially developed for military applications during II world war, that was less sensitive to
intentional interference or jamming by third parties.
Spread spectrum technology has blossomed into one of the fundamental building blocks in
current and next-generation wireless systems

Problem of radio transmission

Narrow band can be wiped out due to interference

To disrupt the communication, the adversary needs to do two things,

to detect that a transmission is taking place and

to transmit a jamming signal which is designed to confuse the receiver.


A spread spectrum system is therefore designed to make these tasks as difficult as possible.

Firstly, the transmitted signal should be difficult to detect by an adversary/jammer, i.e.,the

signal should have a low probability of intercept (LPI).

Secondly, the signal should be difficult to disturb with a jamming signal, i.e.,
thetransmitted signal should possess an anti-jamming (AJ) property


Spread the narrow band signal into a broad band to protect against interference

In a digital communication system the primary resources are Bandwidth andPower. The
study of digital communication system deals with efficient utilization ofthese two resources,
but there are situations where it is necessary to sacrifice their efficient utilization in order to
meet certain other design objectives.

For example to provide a form of secure communication (i.e. the transmitted signal is
not easily detected or recognized by unwanted listeners) the bandwidth of the transmitted
signal is increased in excess of the minimum bandwidth necessary to transmit it. This
requirement is catered by a technique known as Spread SpectrumModulation.

The primary advantage of a Spread Spectrum communication system is its

ability to reject Interference whether it be the unintentional or the

The definition of Spread Spectrum modulation may be stated in two parts.

Spread Spectrum is a mean of transmission in which the data sequence occupies a

BW (Bandwidth) in excess of the minimum BW necessary to transmit it.

The Spectrum Spreading is accomplished before transmission through the use of a

code that is independent of the data sequence. The Same code is used in the receiver
to despread the received signal so that the original data sequence may be recovered.
s(t) wide band r(t) wide band b(t) + Noise
b(t) ..... ... . Narrow Wide
Band Band

c(t) n(t) c(t)

Wide band (noise) Wide band

---- Transmitter---- ---- Channel------ --- Receiver--------

Fig: 8.1 spread spectrum technique.

b(t) = Data Sequence to be transmitted (Narrow Band) c(t)

= Wide Band code
s(t) = c(t) * b(t) (wide Band)

Fig: 8.2Spectrum of signal before & after spreading

Code division multiple access

CDMA is achannel access methodutilized by various radio communication


One of the basic concepts in data communication is the idea of allowing several
transmitters to send information simultaneously over a single communication channel. This
allows several users to share a bandwidth of frequencies. This concept is called multiplexing.
CDMA employs spread-spectrum technology and a special coding scheme (where each
transmitter is assigned a code) to allow multiple users to be multiplexed over the same
physical channel. By contrast, time division multiple access (TDMA) divides access by time,
while frequency-division multiple access (FDMA) divides it by frequency. CDMA is a form
of "spread-spectrum" signaling, since the modulated coded signal has a much higher data
bandwidth than the data being communicated.

An analogy to the problem of multiple access is a room (channel) in which people

wish to communicate with each other. To avoid confusion, people could take turns speaking
(time division), speak at different pitches (frequency division), or speak in different
languages (code division). CDMA is analogous to the last example where people speaking
the same language can understand each other, but not other people. Similarly, in radio
CDMA, each group of users is given a shared code. Many codes occupy the same channel,
but only users associated with a particular code can understand each other.

Technical details

CDMA is a spread spectrum multiple access technique. In CDMA a locally generated

code runs at a much higher rate than the data to be transmitted. Data for transmission is
simply logically XOR (exclusive OR) added with the faster code. The figure shows how
spread spectrum signal is generated. The data signal with pulse duration of Tb is XOR added
with the code signal with pulse duration of Tc. (Note: bandwidth is proportional to 1 / T
where T = bit time) Therefore, the bandwidth of the data signal is 1 / Tb and the bandwidth of
the spread spectrum signal is 1 / Tc. Since Tc is much smaller than Tb, the bandwidth of the
spread spectrum signal is much larger than the bandwidth of the original signal.
Fig. 8.3

CDMA uses Direct Sequence spreading, where spreading process isdone by directly
combining the baseband information to high chip rate binary code. The Spreading Factor is
the ratio of the chips (UMTS = 3.84Mchips/s) to baseband information rate. Spreading
factors vary from 4 to 512 in FDD UMTS. Spreading process gain can in expressed in dBs
(Spreading factor 128 = 21dB gain).
Fig. 8.4
Each user in a CDMA system uses a different code to modulate their signal. Choosing
the codes used to modulate the signal is very important in the performance of CDMA
systems. The best performance will occur when there is good separation between the signal
of a desired user and the signals of other users. The separation of the signals is made by
correlating the received signal with the locally generated code of the desired user. If the
signal matches the desired user's code then the correlation function will be high and the
system can extract that signal. If the desired user's code has nothing in common with the
signal the correlation should be as close to zero as possible (thus eliminating the signal); this
is referred to as cross correlation. If the code is correlated with the signal at any time offset
other than zero, the correlation should be as close to zero as possible. This is referred to as
auto-correlation and is used to reject multi-path interference.

Fig. 8.5

Generation of PN sequence:

Shift Shift Shift Output
Register1 Register2 Register3
S0 S3

Logic Circuit

A feedback shift register is said to be Linear when the feed back logic consists of
entirely mod-2-address ( Ex-or gates). In such a case, the zero state is not permitted. The period of

maximum length sequence or m-sequence.

Example1: Consider the linear feed back shift register as shown in fig 2involve
three flip-flops. The input so is equal to the mod-2 sum of S1 and S3. If the initial state of the shift re

100,110,011,011,101,010,001,100 . . . . . .

The output sequence (output S3) is therefore. 00111010 . . . . .

Which repeats itself with period 2 1 = 7 (n=3)

Maximal length codes are commonly used PN codes

In binary shift register, the maximum length sequence is

N = 2 -1

chips, where m is the number of stages of flip-flops in the shift register.


At each clock pulse

(6) Contents of register shifts one bit right.

(7) Contents of required stages are modulo 2 added and fed back to input.

Fig. 8.8 Initial stages of Shift registers1000 Let initial status of shift register be 1000

1 0 0 0
0 1 0 0
0 0 1 0
1 0 0 1
1 1 0 0
0 1 1 0
1 0 1 1
0 1 0 1
1 0 1 0
1 1 0 1
1 1 1 0
1 1 1 1
0 1 1 1
0 0 1 1
0 0 0 1
1 0 0 0
We can see for shift Register of length m=4.
.At each clock the change in state of flip-flop is shown.

Feed back function is modulo two of X3andX4.

After 15 clock pulses the sequence repeats.
Output sequence is

Properties of PN Sequence
Randomness of PN sequence is tested by following properties

u Balance property
v Run length property
w Autocorrelation property

1. Balance property

In each Period of the sequence , number of binary ones differ from binary zeros by
at most one digit .
Consider output of shift register 0 0 0 1 0 0 1 1 0 1 0 1 1 1 1 Seven
zeros and eight ones -meets balance condition.

2. Run length property

Among the runs of ones and zeros in each period, it is desirable that about one half the runs
of each type are of length 1, one- fourth are of length 2 and one-eighth are of length 3 and so-

Consider output of shift register

Number of runs =8

0 0 0 1 0 0 1 1 0 1 01 1 1 1
3 1 2 2 1 1 1 4

{ No. of agreements No. of disagreements in comparison of one full period } Consider

output of shift register for l=1

Yields PN autocorrelation as
Range of PN Sequence Lengths

Length 0f Shift Register, m PN Sequence Length,

7 127
8 255
9 511
10 1023
11 2047
12 4095
13 8191
17 131071
19 524287
A Notion of Spread Spectrum:

An important attribute of Spread Spectrum modulation is that it can provide

protection against externally generated interfacing signals with finite power. Protection
against jamming (interfacing) waveforms is provided by purposely making the information
bearing signal occupy a BW far in excess of the minimum BW necessary to transmit it.
This has the effect of making the transmitted signal a noise like appearance so as to blend
into the background. Therefore Spread Spectrum is a method of camouflaging the
information bearing signal.
b(t) m(t). . r(t) z(t) Tb
Decisi on Device


c(t) n(t) c(t) Threshold=0


Let { bK} denotes a binary data sequence.

{ cK } denotes a PN sequence.

b(t) and c(t) denotes their NRZ polar representation respectively.

The desired modulation is achieved by applying the data signal b(t) and PN signal c(t) to a product

sequence performs the role of a Spreading Code.

For base band transmission, the product signal m(t) represents the transmitted

signal. Therefore m(t) = c(t).b(t)

The received signal r(t) consists of the transmitted signal m(t) plus an additive
interference noise n(t), Hence

r(t) = m(t) + n(t)

= c(t).b(t) + n(t)


a) Data Signal b(t)


0 -1

b)Spreading Code c(t)


0 -1

c)Product signal or base band transmitted signal m(t)

To recover the original message signal b(t), the received signal r(t) is applied to a
demodulator that consists of a multiplier followed by an integrator and a decision device. The
multiplier is supplied with a locally generated PN sequence that is exact replica of that used
in the transmitter. The multiplier output is given by
Z(t) = r(t).c(t)

4.[b(t) * c(t) + n(t)] c(t)

5.c (t).b(t) + c(t).n(t)

The data signal b(t) is multiplied twice by the PN signal c(t), where as unwanted signal
n(t) is multiplied only once. But c (t) = 1, hence the above equation reduces to

Z(t) = b(t) + c(t).n(t)

Now the data component b(t) is narrowband, where as the spurious component c(t)n(t)
is wide band. Hence by applying the multiplier output to a base band (low pass) filter most of
the power in the spurious component c(t)n(t) is filtered out. Thus the effect of the interference
n(t) is thus significantly reduced at the receiver output.

The integration is carried out for the bit interval 0 t T b to provide the sample
value V. Finally, a decision is made by the receiver.

If V > Threshold Value 0, say binary symbol 1

If V < Threshold Value 0, say binary symbol 0

Frequency Hop Spread Spectrum:

In a frequency hop Spread Spectrum technique, the spectrum of data

modulated carrier is widened by changing the carrier frequency in a pseudo random
manner. The type of spread spectrum in which the carrier hops randomly form one
frequency to another is called FrequencyHop (FH) Spread Spectrum.
Since frequency hopping does not covers the entire spread spectrum
instantaneously. We are led to consider the rate at which the hop occurs. Depending
upon this we have two types of frequency hop.

1. Slow frequency hopping:- In which the symbol rate Rs of the MFSK signal is an
integer multiple of the hop rate Rh. That is several symbols are transmitted on each
frequency hop.

2. Fast Frequency hopping:- In which the hop rate Rh is an integral multiple of the
MFSK symbol rate Rs. That is the carrier frequency will hoop several times during
the transmission of one symbol.

A common modulation format for frequency hopping system is that of M-

ary frequency shift keying (MFSK).
Slow frequency hopping:-

Fig. 8.12 a) Shows the block diagram of an FH / MFSK transmitter, which involves

frequency modulation followed by mixing.

The incoming binary data are applied to an M-ary FSK modulator. The resulting
modulated wave and the output from a digital frequency synthesizer are then applied to a
mixer that consists of a multiplier followed by a band pass filter. The filter is designed

to select the sum frequency component resulting from the multiplication process as the
transmitted signal. An k bit segments of a PN sequence drive the frequencysynthesizer,
which enables the carrier frequency to hop over 2 distinct values. Since frequency
synthesizers are unable to maintain phase coherence over successive hops, most frequency
hops spread spectrum communication system use non coherent M-ary modulation system.
Fig 8.12:- Frequency hop spread M-ary Frequency shift keying
In the receiver the frequency hoping is first removed by mixing the received signal
with the output of a local frequency synthesizer that is synchronized with the transmitter. The
resulting output is then band pass filtered and subsequently processed by a non coherent M-
ary FSK demodulator. To implement this M-ary detector, a bank of M non coherent matched
filters, each of which is matched to one of the MFSK tones is used. By selecting the largest
filtered output, the original transmitted signal is estimated.

An individual FH / MFSK tone of shortest duration is referred as a chip. The chip rate
Rc for an FH / MFSK system is defined by

Rc = Max(Rh,Rs)

Where Rh is the hop rate and Rs is Symbol Rate

In a slow rate frequency hopping multiple symbols are transmitted per hop. Hence
each symbol of a slow FH / MFSK signal is a chip. The bit rate R b of theincoming binary
data. The symbol rate Rs of the MFSK signal, the chip rate Rc and the hop rate Rn are related by

Rc = Rs = Rb /k Rh

where k= log2M

Fast frequency hopping:-

A fast FH / MFSK system differs from a slow FH / MFSK system in that there
are multiple hops per m-ary symbol. Hence in a fast FH / MFSK system each hop is a chip.

Fig. illustrates the variation of the frequency of a slow FH/MFSK signal with time for one
complete period of the PN sequence. The period of the PN sequence is 2 -1 = 15. The
FH/MFSK signal has the following parameters:
Number of bits per MFSK symbol K = 2.
Number of MFSK tones M=2 =4

Length of PN segment per hop k=3

Total number of frequency hops 2 =8
Fig. illustrates the variation of the transmitted frequency of a fast FH/MFSK signal with time.
The signal has the following parameters:

Number of bits per MFSK symbol K = 2.

Number of MFSK tones M=2 =4

Length of PN segment per hop k=3

Total number of frequency hops 2 =8