FOURAH BAY COLLEGE

INSTITUTION
ELECTRICAL AND ELECTRONIC
DEPARTMENT ENGINEERING

Pulse Modulation
MODULE TITLE

EENG524
MODULE CODE

ING PROF JONAS A S REDWOOD-SAWYERR

MODULE LECTURER
Mobile : +232 76 670904, +232 88 001019,
+232 25 273401
Email : [email protected],
CONTACTS
[email protected],
[email protected]
LECTURE HOURS/LAB.
3-0-3
PRACTICALS/CREDIT HOURS
ASSESSMENT Examination will account for 70% of the overall
grade whilst continuous assessment will account
for 30%. Question sets at the end of each
lecture must be returned for grading one week
 after receipt of lectures and will account for 10%
of the continuous assessment grade.
OBJECTIVES AND
GENERAL OBJECTIVES:
OUTCOMES
1. To understand the theory of Pulse
Modulation (i.e. Pulse Analogue
modulation and Pulse Code Modulation) in
contrast to other forms of modulation
schemes discussed in earlier years.
2. To understand the theory of Time Division
Multiplexing (TDM) in Pulse modulation
systems while introducing the concepts of
sampling and quantisation.
3. To understand the theory of modulation in
the presence of noise and the
computation of signal-to-quantisation
noise ratio in communications systems.
4. To understand the concepts defining
Information Theory, while discussing the
properties of sources and channels in the
context of the Shannon and Hartely laws
of information.
5. To understand the theory of Information
content, and Average Information or
entropies of sources and channels.
6. To understand how channels are modelled
for analysis and how to calculate the
average information transmission rate
over discrete channels in terms of source
and destination entropy rates.
7. To define channel capacity in terms of
mutual information and Information
 transmission rates.
8. To analyse the fundamental theorems of
Information theory in terms of information
Rate and Channel capacity.
9. To understand aspects of coding to match
sources to channels by making reference
to different coding schemes.
10. To derive the Syndrome vector [S] in
error detection and correction and to
understand the function of the parity
check matrix [H] in the discussion of
transmission efficiency and redundancy.

MODULE DESCRIPTION Pulse Modulation: Pulse Analogue Modulation,

Pulse code Modulation.
Signal to Quantisation Noise: Linear codes.

Information Theory
Information Content, Entropy or Average
Information Content, Entropy Rate, Entropy of
Joint Event, Redundancy & Information
Efficiency or Relative Entropy, Channel
Representation, Transition Probability Matrix,
Noise-free Channel, Channel Probability or Joint
Probability Matrix, Information Transmission
Rate, Channel Capacity C = Max { I (X : Y)}
bits/symbol, where I (X : Y) = Trans-information
or Mutual Information = H (X) – H (X/Y) = H (Y) –
H (Y/X), Channel Capacity and Information
 Transmission Rate, i.e. C = Max {H’(X) –
H’(X/Y)} bits/sec., Hartley – Shannon Law of
Information, Shannon’s Law for Channel
Capacity C = 2W log2 {(S + N)/N}½ = B
log2(1+S/N) bits/sec., Fundamental Theorems of
Information Theory:
R ≤ C … (1);
R > C … (2).

Coding:
Basic Principles, Binary Coding, Fano Coding
Scheme, Huffman’s Coding Scheme, Coding
Efficiency, Algebraic Codes/Block or (n, k) Codes,
Convolution Code, Error Detection and
Correction, the Syndrome (S) = [H][R], Parity
Check matrix, Hamming Codes.
References
1. Ziemer and Tranter. Principles of
Communications. Systems, Modulation
and Noise. Houghton Miffin
2. A Bruce Carlson. Communication Systems.
SUGGESTED An introduction to Signals and Noise in
Electrical Communications. McGraw Hill.
READINGREFERENCE
3. http://webpages.eng.wayne.edu/
TEXTS/MANUALS/WEBSITES ece4700/Lecture%20Notes/
pulsemodulationpart1.pdf
4. https://www.acsce.edu.in/acsce/wp-
content/uploads/2020/03/ADC-module-3-
final.pdf

01 – Pulse Analogue Modulation

LECTURE NUMBER
 LECTURE DURATION 01
(HOURS)
At the end of this lecture the students should be
able to :
SPECIFIC INSTRUCTIONAL 1. Differentiate between the various forms of
OBJECTIVES AND LEARNING pulse modulation
OUTCOMES 2. Apply the basic theoretical expressions to
solve related engineering problems
3. Analyse the concepts of bandwidth and
signal recovery in pulse analogue
modulation schemes

General description

Analogue (i.e. AM, FM, PM )and digital Continuous wave modulation (ASK, FSK,
PSK) schemes transmit information (i.e. the modulating signal) by modifying some
parameter of a high frequency sinusoidal signal or carrier by the modulating
signal. The baseband or modulating signal can be in different formats, e.g.
speech, data, or some frequency division multiplexed signal.

This course will discuss another important class of modulation system which is
formed from the samples of the baseband signal taken at regular intervals to form
a pulse-like waveform. This process of converting a continuously varying
baseband signal into a pulse-like waveform is referred to as PULSE MODULATION.

PULSE MODULATION

1.0 Pulse modulation – concepts and application

As was mentioned above, the process of converting a continuing varying
baseband signal into a pulse waveform is referred to as Pulse Modulation. This
form of modulation can be categorised as follows :

a) Pulse Analogue Modulation. In this case the information corresponding

to the sampled values of the analogue signal is transferred directly to one of
the parameters of the pulse waveform, i.e. its amplitude, duration or
position). However since the chosen characteristic of the pulse waveform
varies in direct sympathy with the analogue modulating signal, the pulse
modulation is still analogue in nature.
b) Pulse Code Modulation (PCM). In this category of modulation, each
sampled value of the baseband signal is converted into a digital codeword
which is then transmitted. There are a number of levels of coding, i.e.
binary, ternary, quaternary, or in general, m-ary signalling, where binary is
the most common form of coding used in this modulation scheme. In this
case the pulse waveform will consist of a sequence of 1’s and 0’s.

Reasons and benefits of sampling and coding

a) It allows several baseband signals to be transmitted in time sequence by

interleaving the sampled values from each baseband source. This process
is referred to as Time Division Multiplexing.
b) The conversion of the sampled values into coded sequences of binary pulse
trains, as in PCM results in greater immunity of the system to noise and
distortion.
Let us now discuss how the factor of high noise immunity is realized in this
modulation scheme.

Consider Figures 1 a and b

 Threshold level of comparator

1 0 0 1
Threshold level
1 0 0 1 Fig.. 1b
Fig. 1a

The figure demonstrates the process that leads to a minimisation of the noise
pollution in the binary codeword transmitted under this scheme. Fig. 1a shows a
binary codeword 1 0 0 1 which his transmitted. At the receiver due to the
imperfections of the channel in terms of noise, the signal received will be
distorted as seen in Fig1b. The receiver however only has to make a decision
between two levels, i.e. 1 or 0 and then convert back to the original sampled
values. Using a comparator or a Schmitt Trigger with properly set thresholds, this
can be readily done thereby cleaning the corrupted signal or codeword received.

Forms of Pulse Analogue Modulation

In this scheme the analogue baseband signal is made to vary one of the
parameters of a pulse waveform thereby producing a pulse-like waveform which
will contain the baseband information. Given three characteristics of a pulse
waveform, one will therefore expect three types of modulation schemes under
this category:
a) Pulse Amplitude Modulation (P A M). In this format, the height or amplitude
of the pulse is made to vary as a function of the sampled value of the
modulating signal, i.e. the height of the pulse is made proportional to the
sampled value of the baseband signal.
b) Pulse Duration Modulation (P D M) or Pulse Width Modulation (PWM). Here
the width or duration of the pulse varies as a function of the sampled value
 c) Pulse Position Modulation (P P M). In this scheme, the position of the pulse
is made to vary as a function of the sampled value with reference to a
standard position.

Figure 2 demonstrates these various forms of Pulse Analogue Modulation

m(t) Analogue baseband signal

 Sampling instants

g(t)
Gating function g(t). Rectangular
waveform

T t
vPAM(t)
τ
PAM signal. v 0 ∝ m(t)

vPDM(t) τ
PDM signal. τ ∝ m(t)

t
vPPM(t)

PPM signal. t 0 ∝ m(t)

T
to

The PPM and PDM waveforms are referred to as Pulse Time Modulation signals .

GENERATION OF PAM, PDM AND PPM

 1. PAM. Fig.3 provides a simple method of generating a Pulse Amplitude
Modulation signal from an analogue baseband signal using a sampling
gate. Here the sampling gate is basically a transmission circuit in which
the output is a replica of the input waveform during a selected time interval
and zero otherwise. The time interval for transmission is selected by an
externally impressed gating signal, which is normally rectangular in shape.

m(t)
Figure 3.

Analogue vPAM(t) t
Baseband
signal m(t) g(t) τ
V
Gating waveform
g(t) t
T

vPAM(t) τ

t
T
A mechanical analogue of the gating process can be seen in Figure 4
Figure 4
1
revs/ sec
T
m(t) vPAM(t)

τ
θ=
T
There are several forms of gating circuits in the literature using diodes or
transistors. Two of such circuits are shown in Figures 5 & 6.

Referring to Circuit Fig. 5(d) we note that when the gating voltage g(t) is –v 2, both
diodes will be switched off and there is therefore no transmission of the signal
m(t). However when the gating voltage is +v 1, both diodes will be forward biased
and conducting and the output voltage across R L is due only to m(t). It is
important to note that no component of gating voltage will appear across R L due
to the symmetry of the circuit.

Sc(t)
τ

t
a. Sampling circuit with Sc(t) as the gating function
T b. Gating waveform for
Cct a

g(t) vPAM(t)

v1

-v2 t

T
c. Gating waveform for Circuit d
g(t) - g(t)
m(t)

d. Gating
Circuit b
Figure 5.

In Figure 6b, the output voltage equals the input voltage except when the
multivibrator forward biases the diodes thereby clamping the output to zero. This
operation is referred to as Single-ended or unipolar chopping. The trigger
waveform of pulses or switching function S c(t)is shown in Fig.5b

X1 xs(t) m(t)
m(t)
fs
SPDT_OPEN

a. Unipolar Chopper (Switching sampler)

x(t)
τ

t
T
x(t)m(t) v PAM(t)
m(t)
τ

x(t)
t
T

b. Circuit realization c. Waveforms

Fig 6. Unipolar Chopper

Let us now look at the simulation of this process

EWB simulation (sampling gating circuit.ewb)

Fig. 7
Here the input signal is a 100 Hz sinusoid and the clock frequency is 630 Hz. The
output is coloured red showing the PAM waveform at the sampling instants.

Another setting results are shown in Fig. 8

The baseband is a sinusoid with the same frequency as in Fig.7 However the
gating frequency has been changed to 1.271 kHz
Fig. 8

Example 2
PAM.

ii. PDM : In PDM systems the width or duration of the pulse waveform is made to
vary in sympathy with the sampled values of the baseband signal. This is
illustrated in Figure 9. The sampling pulse waveform g(t) in this case must be a
saw-tooth waveform.
m(t)
m(t)

g(t)+m(t) vPDM(t)
t
g(t)
g(t)

g(t)+ Threshold of Schmitt trigger

m(t)

t
vPDM(t)
Figure 9. Generation of a PDM signal.
τ
 t

iii. PPM. A PPM signal can be obtained by first deriving a PDM signal using the
method discussed earlier. From Figure 10, it can be observed that the leading
edge of the PDM waveform is also a function of the sampled value of the
baseband signal. This edge can be used as a trigger input to a Monostable
multivibrator whose output is the PPM signal.

m(t)

g(t)

t
T
τ
vPDM(t)
t

v 1(t)

t
vPDM(t) v1 v2 vPPM(t) v2(t)

v PPM(t)

NB. The diode clips the negative spikes of v 1 and

clamps the positive spikes to ground.
t
PPM signal. t 0 ∝ m(t)
Figure 10. Generation of a PPM signal. to

Spectrum of a PAM signal

We will now consider the PAM signal in greater details due to its importance in
pulse modulation schemes. It forms the basis of the PTM signals as well as the
Pulse Code Modulation scheme since the binary stream of data constituting PCM is
generated from the PAM signal that has been ‘quantised’ and encoded.

Figure 11 shows the pulse waveform of a PAM signal. One can observe that the
PAM signal can be seen to be the product of a switching or sampling function S(t)
and the analogue baseband signal m(t), i.e.

vPAM(t) = S(t)m(t) ...(1.0)

vPAM(t)
 τ

t
T

=
m(t)
S(t) τ
1.0
X
t t
T
Figure 11. Analysis of a PAM signal

From Fourier analysis we can write the sampling function S(t) as

nπτ
∞ sin ( )
τ 2τ T 2nπt
S ( t )= + ∑ cos ( )
T T n=1 nπτ T
( )
T
...(2.0)
∞
τ 2τ nπτ 2 nπt
S ( t )= + ∑ Sinc( )cos ( )
T T n=1 T T
.(3.0)
Let us now inspect this equation and define its features.

T is referred to as the Sampling Interval and fs = 1/T is the Sampling Rate. τ is

called the sampling time.
Let us now discuss the amplitude spectrum of the switching function.

Spectrum of S(t)
τ/T

Sincx curve

0
f
2 −1 1 2
τ τ τ τ
-3fs -2fs -fs fs 2fs 3f ...

Figure 12. Spectrum of the switching function S(t)

NB.
a. We note that the amplitudes of the spectral lines depend on the ratio of the
sampling time to the sampling interval, i.e. τ/T, or the duty cycle of the
waveform.
b. The spectral lines separation depends on the sampling interval T or the
sampling rate fs = 1/T. The lower the sampling rate (i.e. the higher the
value of T), the smaller the separation between adjacent spectral lines.

Spectrum of the PAM signal VPAM(t)

Let us now consider the spectrum of the PAM signal. The expression for the PAM
signal can be obtained by substituting Eqn, (3) in Eqn. (1) to obtain :

( )
∞
τ 2τ nπτ 2nπt
v PAM ( t )= m(t)+ ∑ Sinc m ( t ) . cos ( )
T T n=1 T T
...(4)
Spectra of S(t) and VPAM(t)

The spectrum of S(t) is shown in Figure 12.

The spectrum of S(t) follows the sinc(x) curve shape. The analysis of its features
follows the theory discussed under Fourier series analysis and the plotting of
waveform spectra of continuous signals. The determination of the zero crossing
and the harmonic separation procedures are also applicable. The concept of
reciprocal spreading can be seen in this case as mentioned earlier.

Spectrum of vPAM(t)

Inspecting Eqns 3 and 4 we note that the PAM waveform is a product of the terms
in S(t) and the spectrum of the baseband signal m(t), resulting in a DSBSC single
signal centred on each spectral line of S(t). When the spectrum of m(t) is known
the spectrum for the PAM signal can then be plotted.
From Figure 13a we assume a spectral shape for m(t) and the resulting v PAM(t)
spectrum is shown in Figure 13b.

(a) Spectrum of the baseband m(t) Spectrum of m(t)

 f
-B B

Spectrum of vPAM(t)

Sinc x curve (main

lobe)

-3fs -2fs-B -2fs -2fs+B -fs-B -fs -fs+B -B0 B fs-B fs fs+B 2fs-B 2fs
2fs+B 3fs f
-3/T -2/T -1/T 0 1/T 2/T 3/T

(b) Spectrum of the PAM signal

Figure 13. Spectra of m(t) and vPAM(t)

Demodulation of the PAM signal and the Sampling Theorems

Let us now refer to the spectrum shown in Fig.13. We can see that we have
assumed that that fs – B > B or fs > 2B, i.e. the sampling rate is higher than twice
the highest baseband frequency. Increasing fs further has the effect of increasing
the separation between adjacent spectra. From a practical point, this is desirable
to the transmission engineer since the lower the sampling rate fs (i.e. increasing
the sampling interval T), the more channels or baseband signals can be
interleaved and transmitted in time sequence along the same channel to achieve
multiplex operation. However, lowering the sampling rate has very wide
implications on one’s ability to retrieve the original baseband signal faithfully from
the PAM signal. We shall be advancing this argument later.

If the sampling rate is too low (i.e. a high value of the sampling interval), the
baseband signal may change radically between sampling instants, resulting in loss
of information. We therefore need to find the minimum sampling rate given a
band-limited baseband signal.
Referring to Fig. 13 we can observe that provided fs ≥ 2B, any of the individual
spectra centred on ±nfs, (where n = 0, 1, 2, 3, etc) can be filtered out to obtain
the original baseband signal. A simple method to achieve this is to use a low pass
filter which passes the spectral components centred on the ‘zero frequency’ or DC
component but cuts off sharply before reaching the spectra centred on ±fs.
Fig.14 shows how this can be achieved.
Spectrum of vPAM(t)

Filtered output Sincx

curve

vPAM(t)

m(t) (the original

baseband signal)
a. Block schematic -2fs -fs -B 0 B fs 2fs
-fs+B fs-B
Figure 14. b. Demodulation of PAM
The above criterion namely fs ≥ 2B can be stated in the form of a Sampling
Theorem as follows:

If a signal band-limited to B Hz is sampled at regular intervals and at a

rate fs ≥ 2B, then the samples contain all the information necessary for
the recovery of the original baseband signal.

This is sometimes referred to as the Nyquist Sampling Theorem and the critical
sampling rate, i.e. 2B Hz is called the Nyquist rate.

There is a more generalised form of this statement which we shall now discuss as
Theorem 2
.
Theorem 2

If a signal v(t) is bandlimited between fL and fH, then the minimum sampling rate is
given by :

fs = 2(fH – fL)
if either fH or fL is an harmonic of fs or

2f H
f s=
m
where m is the largest integer not exceeding

fH
( f H −f L )
NB. It must be noted that all higher sampling rates are not necessarily usable
unless they exceed 2fH
Example. Let fH = 60 kHz and fL = 20 kHz. Find fs.
Solution

fH – fL = (60 -20) kHz = 40 kHz. Hence fs = 2 x 40 kHz = 80 kHz.

Neither fH nor fL is a harmonic of fs. Therefore fs ≠ 80 kHz.

Let us now apply the second condition, i.e. fs = 2f H/m, where

m ≤¿ ¿
i.e. Here m ≤ 60/40 = 1.5. From the condition given in the theorem, m = 1 since
it must be the largest integer not exceeding 1.5.

Hence fs = 2 x 60 = 120 kHz.

TIME DIVISION MULTIPLEX (TDM)

One of the basic features of pulse modulation systems is the possibility of

transmitting the sampled values of baseband signals in time sequence over the
same communication channel. The resulting process is referred to as Time
Division Multiplex.

Figure 15 shows a schematic of the PAM/TDM system using the mechanical

analogue representation.

The schematic consists of a transmitting path, at each end of which is connected

a rotating switch, so that the sending and receiving apparatus of each baseband
channel are connected together during their allotted time interval and
disconnected throughout the remainder of the revolution of the switch. In a
practical electronic system the rotating switch could be a sequential switching or
gating circuit.

Baseband m1(t)
m1(t)

m2(t) Multiplex signal

m2(t)

vPAM(t)
m3(t)
Transmitting path
m3(t)

m4(t) m4(t)

Rotating Switch Rotating switch

Figure 15. Mechanical analogue of the TDM process
NB. Switches are synchronised and rotate at 1/T revolutions per second.
The form of the PAM signal which is transmitted is demonstrated in Figure 16. It
consists of interlaced PAM waveforms of identical repetition frequency.
vPAM(t)
Sampling Rate fs = 1/T

t
T
Figure 16. PAM Multiplex signal

NB.
1. It is assumed in the above analysis that the individual baseband signals are
bandlimited and further that the sampling rate fs = 1/T is the same for each
baseband signal.
2. If the original baseband signals are to be recovered by low pass filters in the
receiving system (refer to Figure 14), then from the Sampling Theorem we
require the sampling rate fs to be greater than or equal to the Nyquist rate
demanded for each baseband signal, i.e. the baseband signal which has
the highest frequency component determines the sampling rate fs
of the multiplex system.

Let us now discuss an example to illustrate what has been learnt so far about
TDM.

Example
Five bandlimited baseband signals are sampled and time-division-multiplexed.
The baseband frequencies of the five signals are as follows :

a) 0 – 2.4 kHz
b) 0 – 3.6 kHz
c) 0 – 4.8 kHz
d) 0 – 6.2 kHz
e) 0 – 5.2 kHz
Determine : a) the minimum sampling rate. (b) the maximum sampling time, i.e.
the maximum width of the pulses in the multiplex waveform.
Solution
Part a.
The Nyquist rate for each of the baseband signals is as follows :
a) 2 x 2.4 kHz = 4.8 kHz
b) 2 x 3.6 kHz = 7.2kHz
c) 2 x 4.8 kHz = 9.6 kHz
d) 2 x 6.2 kHz = 12.4 kHz
e) 2 x 5.2 kHz = 10.4 kHz
Hence the minimum sampling rate for the multiplexed system is

fsmin = Highest Nyquist rate of the baseband signals = 12.4 kHz

Part b. The Figure below shows the TDM signal.

vPAM(t)
 τ
T
T
Figure 17. TDM waveform for Problem

In this question we require

1
T=
f smin

Referring to the figure we note that the sampling time must satisfy the following
condition :
T
τ≤
No . of baseband signals
i.e.
T
τ≤
5

or the maximum sampling time is given by

T 1 1
τ max= = = =16.13 μs
5 5 f smin 5 x 12.4 x 10 3

ASSIGNMENT 1

1. Determine the minimum sampling rate required to transmit the following

signals as a PAM waveform.
 i. 0 – 4 kHz signal
ii. 200 – 205 MHz band-limited signal
iii. 3.9 – 8.9 kHz band-limited signal
2. A signal bandlimited between 100 to 101 MHz has the frequency spectrum
shown in Fig Q2 and is sampled at the minimum rate to produce a PAM
signal. Sketch the frequency spectrum of the PAM signal between 0 to 5
MHz. What is the bandwidth of a low pass filter which may be used to
recover the original signal.

Spectrum of baseband signal

100 101 f (MHz)

Paper form submitted

to the Secretary, One
MODE AND DEADLINE FOR SUBMISSION OF week from date of
ASSIGNMENT posting