1

Unit 118 Communication Systems 4]

1.3

Digital Communication and Analogue

Communication.

ANALOG AND DIGITAL MESSAGES

Messages are digital or analog. Digital messages are constructed with a finite number of
symbols. For example, printed language consists of 26 letters, 10 numbers, a space, and
several punctuation marks. Thus, a text is a digital message constructed from about 50
symbols. Human speech is also a digital message, because it is made up from a finite
vocabulary in a language.* Similarly, a Morse-coded telegraph message is a digital message
constructed from a set of only two symbolsmark and space. It is therefore a binary
message, implying only two symbols. A digital message constructed with M symbols is called
an M-ary message.
Analog messages, on the other hand, are characterized by data whose values vary over a
continuous range. For example, the temperature or the atmospheric pressure of a certain
location can vary over a continuous range and can assume an infinite number of possible
values. Similarly, a speech waveform has amplitudes that vary over a continuous range. Over
a given time interval, an infinite number of possible different speech waveforms exist, in
contrast to only a finite number of possible digital messages.

Noise Immunity of Digital Signals

Digital messages are transmitted by using a finite set of electrical waveforms. For example, in
the Morse code, a mark can be transmitted by an electrical pulse of amplitude A/2, and a
space can be transmitted by a pulse of amplitude A/2. In an M-ary case, M distinct electrical
pulses (or waveforms) are used; each of the M pulses represents one of the M possible
symbols. The task of the receiver is to extract a message from a distorted and noisy signal at
the channel output. Message extraction is often easier from digital signals than from analog
signals. Consider a binary case: Two symbols are encoded as rectangular pulses of amplitudes
A/2 and A/2. The only decision at the receiver is the selection between two possible pulses
received, not the details of the pulse shape. The decision is readily made with reasonable
certainty even if the pulses are distorted and noisy (Fig. 1.3). The digital message in Fig. 1.3a
is distorted by the channel, as shown in Fig. 1.3b. Yet, if the distortion is within limits, we can
recover the data without error because we need only to make a simple binary decision as to
whether the received pulse is positive or negative. Figure 1.3c shows the same data with
channel distortion and noise. Here again, the data can be recovered correctly as long as the
distortion and the noise are within limits. In contrast, the waveform in an analog message is
important, and even a slight distortion or interference in the waveform will cause an error in
the received signal. Clearly, a digital communication system is more rugged than an analog
communication system in the sense that it can better withstand noise and distortion (as long
as they are within a limit).

2
Unit 118 Communication Systems 4]
* Here we imply the printed text of the speech rather than its details such as the pronunciation of words and varying inflections,
piteh, emphasis, and so on. The speech signal from a microphone contains all these details. This signal is an analog signal, and its
information content is more than a thousand times the information in the written text of the same speech

A
B
C
D

Figure 1.3 (A) Transmitted signal, (B) Received distorted signal (without noise),(C) Received
distorted signal (with noise),(D) Regenerated signal (delayed).

Viability of Regenerative Repeaters in Digital Communication

The main reason for the superiority of digital systems over analog ones is the viability of
regenerative repeaters in the former. Repeater stations are placed along the communication
path of a digital system at distances short enough to ensure that noise and distortion remain
within a limit. This allows pulse detection with high accuracy. At each repeater station the
incoming pulses are detected and new clean pulses are transmitted to the next repeater
station. This process prevents the accumulation of noise and distortion along the path by
cleaning the pulses periodically at the repeater stations. We can thus transmit messages over
longer distances with greater accuracy. For analog systems, there is no way to avoid
accumulation of noise and distortion along the path. As a result, the distortion and the noise
interference are cumulative over the entire transmission path. To compound the difficulty, the
signal is attenuated continuously over the transmission path. Thus, with increasing distance
the signal becomes weaker, whereas the distortion and the noise become stronger.
Ultimately, the signal, overwhelmed by the distortion and noise, is mutilated. Amplification is
of little help, because it enhances the signal and the noise in the same proportion.
Consequently, the distance over which an analog message can be transmitted is limited by
the transmitter power. Despite these problems, analog communication was used widely and
successfully in the past. Because of the advent of optical fiber and the dramatic cost

3
Unit 118 Communication Systems 4]
reduction achieved in the fabrication of digital circuitry, almost all new communication
systems being installed are digital. But the old analog communication facilities are also in use.

Analog-to-Digital (A/D) Conversion

A meeting ground exists for analog and digital signals: conversion of analog signals to digital
signals (A/D conversion). The frequency spectrum of a signal indicates relative magnitudes of
various frequency components. The sampling theorem (to be proved in Chapter 6) states that
if the highest frequency in the signal spectrum is B (in hertz), the signal can be reconstructed
from its samples, taken at a rate not less than 2B samples per second. This means that in
order to transmit the information in a continuous-time signal, we need only transmit its
samples (Fig. 1.4). Unfortunately, the sample values are still not digital because they lie in a
continuous range and can take on any one of the infinite values in the range. We are back
where we started! This difficulty is neatly resolved by what is known as quantization, where
each sample is approximated, or "rounded off," to the nearest quantized level, as shown in
Fig. 1.4. Amplitudes of the signal m(t) lie in the range (-m p, mp,), which is partitioned into L
intervals, each of magnitude Av = 2mp/L. Each sample amplitude is approximated to the
midpoint of the interval in which the sample value falls. Each sample is now approximated to
one of the L numbers. The information is thus digitized.
The quantized signal is an approximation of the original one. We can improve the accuracy of
the quantized signal to any desired degree by increasing the number of levels L.

For intelligibility of voice signals, for example, L = 8 or 16 is sufficient. For commercial use, L
= 32 is a minimum, and for telephone communication, L = 128 or 256 is commonly used.
During each sampling interval, we transmit one quantized sample, which takes on one of the L
values. This requires L distinct waveforms, which may be constructed, for example, by using a

4
Unit 118 Communication Systems 4]
basic rectangular pulse of amplitude A/2 and its multiples (for instance, A/2, 3 A/2,
5A/2,..., [(L - l)A/2], as shown in Fig. 1.5) to form L distinct waveforms to be assigned to the
L values to be transmitted. Amplitudes of any two of these waveforms are separated by at
least A to guard against noise interference and channel distortion. Another possibility is to use
fewer than L waveforms and form their combinations (codes) to yield L distinct patterns. As an
example, for the case L = 16 we may use 16 pulses (A/2, 3A/2,..., 15A/2, as shown in Fig.
1.5). The second alternative is to use combinations of only two basic pulses, A/2 and -A/2. A
sequence of four such pulses gives 2x2x2x2= 16 distinct patterns, as shown in Fig. 1.6. We
can assign one pattern to each of the 16 quantized values to be transmitted. Each quantized
sample is now coded into a sequence of four binary pulses. This is the so-called binary case,
where signaling is carried out by means of only two basic pulses (or symbols).*
The binary case is of great practical importance because of its simplicity and ease of
detection. Virtually all digital communication today is binary. This scheme of transmitting data
by digitizing and then using pulse codes to transmit the digitized data is known as pulse-code
modulation (PCM).
Atypical distorted binary signal with noise acquired over the channel is shown in Fig. 1.3. If A
is sufficiently large compared to typical noise amplitudes, the receiver can still correctly
distinguish between the two pulses. The pulse amplitude is typically 5 to 10 times the rms
noise amplitude.
Figure 1.5 Multiamplitude pulse code that uses L amplitude levels.

Figure 1.6

Example of a binary pulse code.

5
Unit 118 Communication Systems 4]

For such a high SNR, the probability of error at the receiver is less than 10~6; that is, on the
average, the receiver will make less than one error per million pulses. The effect of random
channel noise and distortion is thus practically eliminated. Hence, when analog signals are
transmitted by digital means, the only error, or uncertainty, in the received signal is that
caused by quantization. By increasing L, we can reduce the uncertainty, or error, caused by
quantization to any desired amount. At the same time, because of the use of regenerative
repeaters, we can transmit signals over a much longer distance than would have been
possible for the analog signal. As will be seen later in this chapter, the price for all these
benefits of digital communication is paid in terms of increased bandwidth of transmission.
From all this discussion, we arrive at a rather interesting (and by no means obvious)
conclusionthat every possible communication can be carried on with a minimum of two
symbols. Thus, merely by using a proper sequence of a wink of the eye, one can convey any
message, be it a conversation, a book, a movie, or an opera star's singing. Every possible
detail (such as various shades of colors of the objects and tones of the voice, etc.) that is
reproducible on a movie screen or on the best quality color television can be conveyed with
no less accuracy, merely by a wink of an eye.*
Although PCM was invented by P. M. Rainey in 1926 and rediscovered by A. H. Reeves in 1939,
it was not until the early sixties that Bell Laboratories installed the first communication link
using PCM. The cost and size of vacuum tube circuits were the chief impediments to the use
of PCM in the early days. It was the transistor that made PCM practicable.

What are the Pros and Cons?

Each paradigm has its own benefits and problems.
Analog

6
Unit 118 Communication Systems 4]
Analog systems are less tolerant to noise, make good use of bandwidth, and are easy to
manipulate mathematically. However, analog signals require hardware receivers and
transmitters that are designed to perfectly fit the particular transmission. If you are working
on a new system, and you decide to change your analog signal, you need to completely
change your transmitters and receivers.
Digital
Digital signals are more tolerant to noise, but digital signals can be completely corrupted in
the presence of excess noise. In digital signals, noise could cause a 1 to be interpreted as a 0
and vice versa, which makes the received data different than the original data. Imagine if the
army transmitted a position coordinate to a missile digitally, and a single bit was received in
error? This single bit error could cause a missile to miss its target by miles. Luckily, there are
systems in place to prevent this sort of scenario, such as checksums and CRCs, which tell the
receiver when a bit has been corrupted and ask the transmitter to resend the data. The
primary benefit of digital signals is that they can be handled by simple, standardized receivers
and transmitters, and the signal can be then dealt with in software (which is comparatively
cheap to change).
* Of course, to convey the information in a movie or a television program in real time, the winking would have to
be at an inhumanly high speed.

The advantages of digital signals over analog signals

1. Noise and distortion immunity:-when analog signal is transmitted in free space for long

2.
3.
4.
5.
6.
7.

distances it will be distorted and no an ideal recovery done in receiver to recover the
exact signal which transmitted ,but in digital communication digital signal consist of 2
levels only zero or one (not like the analog one , it has infinite numbers of values or
levels at each period of time),so digital signal can be recovered easily and exactly like
transmitted one. Device that used to partially recover the analog signal after a certain
distance is called repeater ,from its name it receives transmitted analog signal and
filter it and amplify it and transmit it),but in digital transmission it is different ,the
device used for recovering digital signal which transmitted from a scertain distance is
called regenerative repeaters ,it receives transmitted digital signal and recover it to
regenerate a new exact signal like received one then retransmit it for further distance.
More privacy and security due to encryption.
Error detection techniques can be used and correction as well.
Easy for processing and applying multiplexing techniques.
Low cost.
Transmission rate can be changed easily.
Digital hardware implementation is flexible and permits the use of micrprocessors,
digital switching elements and layer scale.