Review

of
Signals & Systems
PII
 Signal and Systems

● Electrical communication signals are time varying quantities such as voltage or

current.

● A system is any process that produces an output signal in response to an input

signal. A signal physically exists in the time domain, it can also be represented in
frequency domain where the signals are viewed as consisting of sinusoidal
component at various frequencies.

● Converting the time domain to frequency domain requires the use of Fourier Series
and Fourier transforms
Signal and Systems

● Classification of Signals
We consider one-dimensional signals defined as single-valued functions of time.
“Single-valued” means that for every instant of time there is a unique value of the
function.
1. Continuous –time Signal and Discrete-time Signals

A signal is a continuous-time signal if it is defined for all time �. Continuous-time signals

arise naturally when a physical waveform such as an acoustic wave or light wave is
converted into an electrical signal. The conversion is effected by means of a transducer;
A discrete-time signal is defined only at discrete instant of times. Thus, in this case, the
independent variable has discrete values only, which are usually uniformly spaced. A
discrete-time signal is often derived from a continuous-time signal by sampling it at a
uniform rate.
2. Periodic and Non-periodic

A signal �(�) is periodic if �(�) = �(� + ��), ∀ �, ��� � ∈ ℤ

�[�] = �[� + �]
The smallest value of � > 0 that satisfies the equation known as the fundamental period
of �(�).
 Signal and Systems

● A signal �(�) is aperiodic or non-periodic if �(�) ≠ �(� + ��), ∀ �, ��� � ∈ ℤ

3. Deterministic and Non-deterministic Signals

q A signal �(�) or �[�] is deterministic if it can be completely specified in time, (it

has no uncertainty with respect to its value at any time), that is, the nature and
amplitude of such signals at any time can be predicted.

q A signal is non-deterministic if there is uncertainty before its actual occurrence. It

has patterns that are quite irregular. It is otherwise refer to as random signal.
Typical example is noise.
Signal and Systems

4. Energy and Power Signals

Energy of Signal
Consider a signal �(�) i.e., � is a function of �
The Energy �� of signal �(�) is defined as
∞ 2
�� = −∞
�(�) �� (2.1)
−�
Example: consider �(�) = � � ≥ 0,
0 �<0
this can also be written as �(�) = �−� �(�), where, �(�) is the unit step function.
∞ 2 ∞ −2�
Therefore, �� = 0
�(�) �� = 0
� ��
1
= − �−2� ∞
0 =1 2
2

�(�) is an ENERGY SIGNAL if �� is FINITE i.e., �� < ∞

Signal and Systems
● Power of Signal
The power �� of signal �(�) is
1 �
�� = lim − �
2 �(�) 2 �� (2.2)
�→∞ � 2

������ �� ������ �� ���� �

= lim
�→∞ �

�(�) is an POWER SIGNAL if �� is FINITE i.e., �� < ∞

1 �2
IF �(�) is an Energy signal, �� = lim −� �(�) 2 ��
�→∞ � 2
1 ∞
Observe that ≤ lim �(�) 2 ��
�→∞ � −∞
��
= lim = 0
�→∞ �

Therefore, �� ≤ 0 from above inequality, and

�� ≥ 0 since �� is non-negative
⟹ �� = 0
Power of Energy Signal is ZERO
Signal and Systems
IF �(�) is a power signal,
Energy in a window of � ≈ �� ∙ �

⟹ Total Energy = lim .Energy in window of size �

�→∞

= lim �� ∙ � = 0
�→∞

Energy of a POWER SIGNAL is ∞

A signal is a POWER SIGNAL if its power is finite, and its energy is Infinite
A signal is a ENERGY SIGNAL if its energy is finite, and its power is zero

The Energy and Power classification of signals are mutually exclusive. Periodic signals
are usually viewed as power signals, whereas non-periodic are energy signals.
Signal and Systems

Exercise 1
1) Classify the following signals as either energy signal or power signal. Hence
find the normalized energy or normalized power of each as the case may be.
a) �(�) = �−3 �
b) �(�) = 2 ��� �

c) �(�) = ��� � + 3 ��� 2�

d) �(�) = �−2�

2) Show that the signal defined as

�(�) = ��(�)
is neither energy nor a power signal
Signal and Systems
Power of a Periodic Signal
Let � = Fundamental Period
1 �
�� = lim − �
2 �(�) 2 ��
�→∞ � 2

chose � = ��, � → ∞ ⟹ �� → ∞
⟹ τ→∞
�� 1
�� = lim ��
2 �(�) 2 ��
�→∞ �� − 2

(energy in m period = � × ������ �� � ������)

1 �
= lim .� 2
−� 2
�(�) 2 ��
�→∞ ��

1 �
�� = − �
2 �(�) 2 ��
� 2

�� = ������ �� ������ �� ���� �

2
Determine the normalized power �� in �(�) = � ��� (2���), Ans = � 2
Signals and Systems
● Singularity Functions
1) The unit step function �(�)
1

1 �>0
�(�) = (2.3)
0 �<0

2) Unit impulse functions �(�)

The unit impulse function also known as the Dirac delta function denoted as
�(�) is the derivative of the unit step.
t
−∞
�(t) = u(t) = 1 (2.4)

It is a limit of a sequence of pulse-like function as they become narrower and

simultaneously taller. It is assumed to have infinite amplitude and zero width. (It is a
theoretical signal and not a function in a strict time mathematical sense).
 Signals and Systems
● Shifting property of Delta function
● The shifting property of delta function states that
� �(�0 ) � < �0 < �
�
�(�)�(� − �0 ) �� = (2.5)
0 ��ℎ������
That is, if we multiply any function �(�) by �(� − �0 ) , and integrate from a to b, we will
obtain the value of �(�) evaluated at � = �0
Odd and Even Function
q A function �(�) is said to be EVEN if �(−�) = �(�).

q A graph of an EVEN function is symmetrical about the y-axis. E.g., �(�) = �2 ; (since
�(−2) = �(2) = 4), �(�) = ��� �, (���(−�) = ��� �)
q A function �(�) is said to be ODD if �(−�) ≠ �(�) but rather, �(−�) = − �(�).

q The graph of an ODD function is symmetrical about the origin. E.g., �(�) = �3 ; �(�) =
��� �,

q For EVEN function, the waveform is symmetrical about the y-axis. For the ODD
function, it is symmetrical about the origin. A waveform that is symmetrical about
the horizontal axis have average area equal to zero.
 Signals and Systems

Classification of Systems
1. Linear and Non-linear System.

A system is said to be linear if it obeys the principle of superposition.

2. Time-Invariant and Time-Varying systems
A system is time invariant if a time shift in the input results in a corresponding time
shifts in the output
3. Casual and Non casual System.

A system is said to be causal if the present value of the output signal depends only on
the present and/or past values of the input signal. In contrast, the output signal of a
non-causal system depends on future values of the input signal.
Signals and Systems

Exercise 2

Verify the linearity or otherwise of the systems defined by the following

1) �(�) = ��(�)
2) �(�) = �2 (�)
Signals and Systems
Line Spectral
A phasor is a rotating vector representation of a sinusoid whose amplitude A, phase �,
and frequency � are time–invariant. Consider the sinusoid given by
�(�) = � ���(�� + �) (2.6)

where, ω = 2��0 = 2� �

This can be represented mathematically as the sum of two complex-valued functions

� ���(�� + �) = � 2 ��(��+�) + � 2 �−�(��+�) (2.7)

and for the one-sided phasor the function is of the form

��[� ���(�0 � + �)] = ���(��+�) (2.8)

since ��[�] = 1 2 [� + �∗ ]

Therefore, the phasor has length A, rotate

anticlockwise at a rate �0 rev/sec and at time � = 0
make angle � with the real axis.
Signals and Systems
Line Spectral
Three parameters therefore completely specify a phasor;
Amplitude, Phase angle, and Rotating frequency

Figure: Two-sided phasor

representation

In frequency domain, amplitude and phase must be associated with the particular
frequency �0 .

A frequency domain description of a periodic signal would be a line spectrum of 2 plots:

1) Amplitude vs Frequency and 2) Phase vs Frequency.
 Signal and Systems - Line Spectra
Four conventions are adhered to regarding the line spectra
1) Independent variable would be cyclical frequency
2) Phase angle would be mesured with respect to cosine wave,
sine wave needs to be converted to cosine via the identity
��� �� = ���(�� − 900 )
3) Amplitude is regarded as always +ve. When -ve sign appears,
it must be absorbed in the phase using −������ = ����(�� ± 1800 ).
It does not matter whether it is +180 or − 1800 .

4) Phase angle ususally are expressed in degrees

Example 1: Draw the one-sided sprectrum for the signal

7 − 10 ���(40�� − 600 ) + 4 ���120��

Solution: Converting the constant term to a zero frequency or DC, and apply the rule (2)
and (3) from above, the expression becomes

7��� 2�(0)� + 10 ���(2�(20)� + 1200 ) + 4 ��� (2�(60)� − 900 )

Signal and Systems
Line Spectral

One-sided phasor representation

For the two-sided representation: recall that

� ���(�� + �) = � 2 ��(��+�) + � 2 �−�(��+�)

�(�) = 7��(0)� + 10 2 ��(2�(20)�+120 ) + 10 2 �−�(2�(20)�+120

0 0 )7
Therefore,

+ 4 2 ��(2�(60)�−90 ) + 4 2 �−�(2�(60)�−90
0 0)
Signals and Systems
Line Spectral

Exercise 3: Draw the double-sided line spectra for the signal

�(�) = 3��� � + � 4 + 4��� 2� − � 8 + 6��� 3�

 Signal and Systems
Frequency Domain Representation Of Signals
Fundamental tool for communication
This can be of two types
q Fourier Series ---- Discrete in frequency
Define for a periodic (Power) signal
q Fourier Transform ---Continuous in Frequency
For non-periodic (Energy) Signal

Fourier Series

Exponential Fourier Trigonometric

Series Fourier Series
Signals and Systems
Exponential Fourier Series
Let �(�) be a power signal with period � = 1 � . It’s exponential Fourier series
�
expansion is
∞
�(�) = � ��2����� ,
�=−∞ �
� � ℤ, i.e., � = . . . , − 2, − 1, 0, 1, 2, . . . (2.9)

�� is ��ℎ discrete Fourier series coefficients and, it is related to �(�) by

1 �
−�2���� �
�� = −
2
� �(�)� �� (2.10)
� 2

The �(�) in () consists of phasors with amplitude �� and angle ��� �� at frequencies
��� = 0, ± �� , ± 2�� , . . .
Three important spectral properties of periodic power signal are:
1) All frequencies are integer multiples or harmonics of the fundamental frequency.
Hence, the spectral lines have uniform spacing �� .

2) The d.c component equals the average value of the signal i.e., �0 = 1 � �
�(�)��

3) If �(�) is a real function of time then, �−� = �∗� = �� �−� ��� �� . This would implies
that the amplitude spectrum has even symmetry and the phase has odd symmetry.
Signals and Systems
and
∞
�(�) = �0 + �=1
2�� ���(2��� � + ��� �� ) (2.11)

Example 2 A

Consider the periodic signal represented

in Figure . Assume � = � 4 . Determine -T-d -T d T T+d
the Fourier coefficients.

Solution: Any periodic signal �(�) can be decomposed into an infinite number of
complex sinusoid as �(�) = ∞ �+−∞
� � � �2���0 �
, where � � = � �ℎ
Fourier coefficient and, �
is the ��ℎ Harmonic. To compute �� , � � ℤ
1 �
�� = 2 �(�)�−�2���0 � ��
�
� − 2

Note: The integer can be over any period of �(�) i.e., from � to � + �
1 � 1 �
∴ �� = �(�)�−�2���0� �� = 4 �. �−�2���0 � ��
� 0 � 0
�
Since signal is non-zero only between 0 and �
� 4 �−�2���0 � ��
4 �� =
� 0
 Signals and Systems
� � � �
When � = 0; �0 = 4 1 �� = . = �4
� 0 � 4

�
∴ �0 = 4 d.c coefficient (i.e., at zero frequency)
When � ≠ 0;
�
1 � � �−�2���0� 4
�� = 4 �. �−�2���0 � �� = .
� 0 � −�2���� 0
�
� �−�2���0 4 −1 � ��
= = 1 − �−� 2
� −�2���0 �2��
��
Factoring out �−� 4 from the expression in bracket, we obtain
�� �� �� ��
2� ��� �� 4
� �
�� = . �−� 4 �� 4 − �−� 4 = . �−� 4.
�2�� �2��

� �� ��
�� = ��� �−� 4 the ��ℎ Fourier series coefficient
�� 4
�
4, �=0
�� = .
� �� −���
��� � 4 , �≠0
�� 4
 Signals and Systems
The magnitude for � ≠ 0 can be expressed as
� �� −���
�� = ��� � 4
�� 4

��
� �� −�
= ��
���
4
since � 4 = 1 (Try and verify this!)

This is known as MAGNITUDE SPECTRUM

Signals and Systems

Example 3:

For the rectangular waveform shown in Figure , obtain the complex exponential Fourier
series representation.
We take the range of integration over the
� �
central period i.e., − ≤ � ≤ where,
2 2
A
� �
�: − <�<
2 2
�(�) = � �
0: � <− , � >
2 2 � �
� �− 2 2 � �
2 2

� �
Thus, �� = 1 � 2 �(�)�−�2���� � �� =
−� 2
1
�
2 �. �−�2���� � ��
−� 2
�
� −�2���� � 2 �
= � = ������� − �−������
−�2���� � −� 2 �2���� �
�
= . 2���� ���� �
�2���� �
 Signals and Systems
� � ��� ���� �
. 2���� ���� � = .
�2���� � � ����

Multiplying and dividing by �, finally gives

�� ��� ���� �
�� =
� ���� �

��
= ���� ��� � = ��� � ���� ��� �
�

The amplitude spectrum obtained from �� = ��� � ���� ��� � is as shown for the case
of � � = �� � = 1 4 .

Therefore the Fourier series of �(�) can then be expressed as

(� � = 1 4, � = 0)
��
�0 = ���� ��� � = � 4
�

��� ��
2�� = 2� 4 ���� � 4 = 2� 4 �� 4 = 2� �� ��� �� 4
4

�(�) = � 4 + 2� � ��� � 4 ����0 � + 2� 2� ��� 2� 4 ���2�0 � + . . .

 Signals and Systems

Figure Spetrum of rectangular pulse train with �� � = � � : (a) Amplitude, (b) Phase

�(�) = � 4 +
√2 � √2
�����0 � + ���2�0 � + ����3�0 � + . . .
� � 3�
Signals and Systems
Parseval’s Power Theorem
This relates the average power of a periodic signal to its Fourier coefficients.
∞ ∗ ∞ 2
i.e., � = �=−∞
� � � � = �=−∞
��

Prove !!!
 Signals and Systems
Trigonometric Fourier Series
The function �(�) that satisfies the conditions earlier stated may be
represented in trigonometric form by the infinite series as:

(2.12)

�0 , �� , and �� are constants which depend on � and �(�). The process of

determining the values of the constants is called Fourier Analysis.
From above, the periodic function consists of sinusoidal components of
frequencies 0, �0 , ..., ��0 . This representation of function �(�) is in the
frequency domain.
q The first component �0 with zero frequency is called d.c component.

q The sine and cosine terms represent the harmonics.

q The number of � represents the order of the harmonics.

 Signals and Systems
Useful Integral calculus
The Fourier coefficients or constants can be evaluated by integration. Hence, the following theorems will be
useful.
2� 1 1
1) 0
��� �� �� = [��� ��]2�
0 = � (1 − 1) = 0
�

2� 1 1
2) 0
��� �� �� = − [��� ��]2�
0 =− � (1 − 1) = 0
�

2� 1 2� 1 1 2� 1
3) 0
���2 �� �� = 0
(1 − ��� 2��)�� = � − 2� ��� 2�� = 2 (2�) = �
2 2 0

2� 1 2� 1 1 2� 1
4) 0
���2 �� �� = 0
(��� 2�� + 1)�� =
2�
��� 2�� + � = 2 (2�) = �
2 2 0

2� 1 2�
5) 0
��� �� ��� �� �� = 0
���(� + �)� + ���(� − �)� ��
2

1 1 1 2�
= − �+� ���(� + �)� − �−� ���(� − �)� =0
2 0

2� 1 2�
6) 0
��� �� ��� �� �� = 0
���(� + �)� + ���(� − �)� ��
2

1 1 1 2�
= 2
− �+� ���(� + �)� − �−� ���(� − �)� =0 :�≠�
0

2� 1 2�
7) 0
��� �� ��� �� �� = 0
���(� − �)� − ���(� + �)� ��
2

1 1 1 2�
= 2 �−�
���(� − �)� − �+� ���(� + �)� =0 :�≠� (�, � � ℤ)
0
 Signals and Systems - Trigonometric Fourier Series
Value of �0

Replacing �� � by � in ()
from

Integrating both sides of the series

2�
0
�(�)�� = �0 [�]2�
0 + 0 + 0 + . . . = 2��0
2�
∴ �0 = 1 2� 0
�(�)��

If we take the periodic function as �(�) and integrate over a period T (which corresponds
to 2�) , we get
1 �+�
�0 = �(�)�� (2.13)
� �
where t is any value.
 Signals and Systems - Trigonometric Fourier Series
Value of ��

To find �� we multiply both sise of the Fourier aeries by ��� �� and integrate between
the limits � = 0 to 2�.
2� 2� ∞ 2�
0
�(�)��� �� �� = �0 0
��� �� �� + �
�=1 � 0
��� �� ��� �� ��
∞ 2�
+ �
�=1 � 0
��� �� ��� �� ��

∞ 2� 0 �≠�
= �
�=1 � 0
��� �� ��� �� �� =
�� � �=�
2�
�� = 1 � 0
�(�)��� �� ��

If we take the periodic function as �(�), then the expression for �� becomes
�+� �+�
�� = 2 � ��� �� � �� = 2 �
2��
�
�(�) �
�(�)��� � �� (2.14)
�

Value of ��
Following the same procedure as above, we can obtain �� as
�+� �+�
�� = 2 � ��� �� � �� = 2 �
2��
�
�(�) �
�(�)��� � �� (2.15)
�
Signals and Systems
Example 4: Determine the trigonometric Fourier series to represent the periodic
waveform shown below.
�(�)
�

0 2� 4� t

Ans: �(�) = � 2 − ��� � − 1 2 ��� 2� − 1 3 ��� 3� − 1 4 ��� 4� − 1 5 ��� 5�

Example 5: Find the Fourier series for the waveform shown below
�(�)
4

-� − � 0 � � 3� 2� �
2 2 2

Ans: �(�) = 2 + 8 � ��� � − 1 3 ��� 3� + 1 5 ��� 5� − 1 7 ��� 7� + . . .

Signals and Systems
Example 6: Find the trigonometric Fourier series representation for the half wave
rectified sine wave where

0 −�<�<0
�(�) =
��� � 0<�<�

Ans:

∞
2
�(�) = 1 � + 1 2 ��� � + ��� ��
(1 − �2 )�
�=����
 Signals and Systems - Trigonometric Fourier Series
Useful theorems
Theorem 1:

If �(�) is defined over the interval −� < � < � and �(�) is EVEN then, the Fourier series
for �(�) contain the constant term �0 and the cosine terms only, i.e., �� = 0.
∞
�(�) = �0 + �
�=1 �
��� �� (2.16)

Theorem 2:

If �(�) is an ODD function defined over the interval −� < � < � then, the Fourier series
for �(�) contain the constant term �0 and the sine terms only, i.e., �� = 0.
∞
�(�) = �0 + �
�=1 �
��� �� (2.17)

Theorem 3:

If �(�) is neither EVEN nor ODD function defined over the interval −� < � < � then, the
Fourier series for �(�) contain the constant term �0 , the cosine term and the sine terms,
i.e.,
∞
�(�) = �0 + �=1
( �� ��� �� + �� ��� ��) (2.18)
 Signals and Systems - Trigonometric Fourier Series
Note:

o The sum or product of two or more EVEN function is an EVEN function, e.g.,

�2 ∗ ��� � = �2 ��� �
o The sum of two or more ODD functions is an ODD function and the product of two
or more ODD functions is an EVEN function

o The product of EVEN function and a ODD function is an ODD function, e.g., �2 ��� �

Exercise 4
The Figure below shows a periodic wave signal which is symmetrical with respect to the
vertical axis. Obtain its Fourier series representation. Using (i) Exponential method (ii)
Trigonometric method.
 Signals and Systems
Fourier Transform and Continuous Spectra

q We have described a periodic signal as one with a repeating characteristic.

q A nonperiodic signal are signals whose effects are concentrated over a brief
period of time. It could be strictly timelimited.
q It is identically zero outside of a specified interval, i.e., �(�) → 0 as � → ∞
q The signal total energy is well defined (finite) and the average power equals
zero

For a periodic signaly, its complex exponential Fourier series can be

represented as
∞ 1 . −�2���� �
�(�) = �=−∞ �0 �0 �(�)� �� ��2����� (2.19)
��
Fourier Transform and Continuous Spectra
Analogously, for a nonperiodic signal, �(�) can be expressed as
∞ ∞
�(�) = −∞ −∞
�(�)�−�2��� �� ��2��� �� (2.20)
�(�)

where the bracketed term is the Fourier transform of �(�) symbolized as �(�) or ℱ[�(�)]
∞
�(�) = ℱ[�(�)] = −∞
�(�)�−�2��� �� (2.21)

It is an integration over all time.

The expression in (2.20) states that �(�) can be found by inverse Fourier transform of
�(�) as
∞
�(�) = ℱ−1 [�(�)] = −∞
�(�)� �2���
�� (2.22)

which is an integration over all frequency

Equations (2.21) and (2.22) are often referred to as the FOURIER INTEGRAL or the
FOURIER TRANSFORM PAIR
Comparing the two Fourier analysis, i.e., F.S and F.T
q (2.19) and (2.20) show that �(�) plays the same role for nonperiodic signals that ��
plays for periodic signals
Fourier Transform and Continuous Spectra
q �(�) is the spectrum of the nonperiodic signal of �(�) while �� is the spectrum of the
periodic signals.
q �(�) is a continous function defined for all values of � whereas, �� is defined only for
discrete frequencies.
q Therefore, a nonperiodic signal will have a continuous spectrum rather than a line
spectrum as in periodic signals.
Example 7:

Obtain the Fourier trsansform of the rectangular pulse train shown below.
Ans: �� ���� �� Fig. Example 7

Example 8:

Determine the Fourier transform of the signal �(�) shown below.

�
���� �� 2 − �−�2��
�
Ans: �−�� 2
A
��

Fig. Example 8
T 2T
Fourier Transform and Continuous Spectra

Important Properties of �(�)

1). If �(�) is real, then �(−�) = �∗ (�) :

�(−�) = �(�) , ���[�(−�)] =− ���[�(�)] (5)
2). If �(�) has either EVEN or ODD time symmetry, then (3) simplifies to
∞
2 �(�)��� �� �� , �(�) ����
0
�(�) = ∞
−�2 �(�) ��� �� �� , �(�) ���
0

3). The value of �(�) at � = 0 equals the net area of �(�) i.e.,
∞
�(0) = −∞
�(�)��

q This compares with the periodic case where �� (0) equals the average value of �(�).
Fourier Transform and Continuous Spectra

Some Properties of Fourier Transform

1) Duality theorem
If we re-examine the pair of Fourier integrals, it can be seen that they differ only by the
variable of integration and the sign in the exponential.
The consequence of this similarity is the duality theorem: i.e., �(�) and �(�) constitute a
known transform pair.

2) Time delay and scale change

Given a time function �(�), various other waveforms can be generated from it by
modifyimg the argument of the function. E.g., replacing � by � − �� produces the time-
delayed signal �(� − �� ).
The delayed signal has the same shape as �(�) but shifted �� units to the right
along the time axis.
In frequency domain, time delay causes an added linear phase with slope −2��� . So that
�(� − �� ) ⟺ �(�)�−�2����

and �(� + �� ) ⟺ �(�)��2����

Fourier Transform and Continuous Spectra

3) Frequency translation
A dual of the time-delay theorem is

�(t)��2���� ⟺ �(� − �� )
This is designated as frequency translation or complex modulation, since multiplying a
time function by �2���� causes its spectrum to be translated in frequency by + ��

Let �(�) have the bandlimited spectrum of �, and let �� ≫ �.

The inspection of the translated spectrum �(� − �� ) reveals the following:
q The significant component are concentrated around the frequency ��
q Though �(�) was bandlimited in �, �(� − �� ) has a spectral width of 2�; translation
has therefore doubled the spectral width. The negative portion of �(�) now appears
at positive frequecies.
MODULATION
PIII
Modulation
q Modulation is the process of imposing information contained in a lower-frequency
electronic signal unto a higher frequency signal. The higher-frequency esignal is
called the carrier and the lower-frequency signal is called the modulating signal
(Baseband).
q It could be referred to as a process that moves the message signal into a specific
frequency band that is dicted by the physical channel.
q It produces a number of benefits as ealier enumerated
q It could be analog or digital.
q Traditional communication systems such as AM / FM radios and PAL / NTSC
television signals are based on analog modulations while, more recent systems such
as 2G, 3G, 4G, 5G cellulars, HDTV etc. are all digitals.
q Baseband is used to designate frequency band of the original message signal from the
source or the input transducer. In TV the video baseband is the video band
occupying 0 to 3.58Mhz / 4.3MHz for PAL / NTSC. For digital data or pulse code
modulation (PCM) that uses bipolar signaling at rate of Rb pulses per second, the
baseband is approximetely 0 to Rb Hz.
 Modulation
q In baseband communication, message signals are directly transmitted without any
modification and, because they contain significantly low-frequency content, they
cannot be effectively transmitted over radio link. Instead, dedicated user channels
such as twisted pairs of copper wires and coaxial cables are used.
q Communication that uses modulation to shift the frequency spectrum of a signal is
known as carrier communication. In analog modulation, one of the basic parameters
of a sinusoidal carrier of high frequency is veried linearly with the baseband signal.
q Carrier Frequencies and Bandwidths

q Although, the parameters B and fc would seem to be unrelated theoretically,

practical factors cause them to be connected, albeit rather loosely.

q As a rough rule-of-thumb, the fractional bandwidth � � must be bounded by

�

0.01 < � � < 0.1

�

otherwise, there will be signal distortion from the system.

 Modulation

Therefore, large bandwidth requires high carrier frequencies.

See table below for some selected carrier frequencies and nominal bandwidth

Observation shows that � ≈ 0.02��

Modulation
Consider a sinusoidal signal

�(�) = �(�)���[�� � + �(�)] (3.1)

Three variables can be identified
1). Amplitude 2) Instantaneous frequency 3) Phase

The message �(�) can be used to modulate any one of these three parameters to
allow �(�) to carry the information from the transmitter to the receiver.
It can result in either:
Amplitude modulation -------- when amplitude varies with �(�)
Frequency modulation --------- when the frequency varies with �(�)
Phase modulation ----------- when the phase varies with �(�)
Amplitude modulation is linear while, the other two are similar and nonlinear, and
are referred to as angle modulation.
 Linear Modulation
Amplitude Modulation (AM)

The unique feature of AM is that the envelope of the modulated carrier has the same
shape as the message waveform.

This is achieved by adding the translated message to thr unmodulated carrier.

��
Consider, an unmodulated carrier of �� ����� � where, �� = 2� and a modulating
signal designated as �(�). The modulated signal becomes

�� (�) = �� ��� �� � + ��(�)�� ����� � = �� [1 + ��(�)]����� � (3.2)

where, � is the modulation index.

�� (�) = �� [1 + ��(�)]����� � is the envelope of the modulated carrier. It is a linear

function of the message hence, the designated as amplitude modulated
Linear Modulation
 Linear Modulation
Amplitude Modulation

The envelope of the modulated carrier has the shape of �(�) provided that:
1) �� ≫ �� 2) �≤1
m � = 1 ⟹ 100% modulation, � > 1 ⟹ overmodulation
Note: for easy signal recovery using simple demodulation scheme, the signal amplitude
has to be small and the DC component of the signal has to be zero, that is,
.1
�(�) < 1, and lim �(�)�� =0
�→∞ � �

In the frequency domain, the spectrum of the AM signal is given by

�� ���
�� (�) = [�(� − �� ) + �(� + �� )] + [�(� − �� ) + �(� + �� )] (3.3)
2 2

The spectrum consists of the translated message plus a pair of impulses at ± ��

representing the carrier itself.
 Linear Modulation

Figure: Amplitude modulation. (a) Sinusoidal message signal. (b) AM signal.

(c) Message spectrum for an arbitrary x(t). (d) Modulated signal spectrum.
 Linear Modulation
Amplitude Modulation
The modulation index can be defined as
�(�) ��� − �(�) ���
�= (3.4)
�(�) ��� + �(�) ���

Transmit Power
Let the unmodulated carrier be represented as ����2��� � and if we assume a single
audio tone as the modulating signal such that the message signal be represented as
� ���2��� �.
The modulated wave can be represented mathematically as
�� (�) = (� + ����2��� �)(���2��� �)

= � 1 + � � ���2��� � ��� 2��� � ⟹ �� [1 + ��(�)]����� �

if we use the identity ���� ���� = 1 2 ���(� − �) − 1 2 ���(� + �), then

�� ��
�� (�) = ���� 2��� �+ ��� 2�(�� − �� )� − ��� 2�(�� − �� )� (3.5)
2 2
 Linear Modulation
Since power is proportional to the square of voltage / amplitude, then
�� 2 �� 2
�� = + + �2
2 2

� 2 �2 � 2 �2 2 �2 �� �2 ��
= + +� = + + ��
4 4 4 4

where, �� is the power content of the carrier and, it is independent of the modulation
index in the AM transmission.
�2 �� � 2
�� = + �� = �� 1 + 2 (3.6)
2

Under optimum condition (� = 1), only one-third of the power transmitted is located in
the sidebands. Two-third of the power is located at the carrier frequency.
Note:

q No information is contained at the carrier frequency

q All information is contained within the upper and lower sidebands

q The two sidebands contain identical information

q The simplicity of the AM depends on the power in the carrier (detail later) hence, the
carrier is not altogether a waste.
 Linear Modulation
Amplitude Modulation
In order to make the better use of the available power being transmitted, other more
efficient schemes such as DSB-SC, SSB and VSB were introduced. In each of this
schemes, power is put where the information is.