Analog and digital signals

Dr.R.Subasri
Professor, EIE
Kongu Engineering College
Signal is a way of communicating by sending information from one system to other system. Signal is an
electromagnetic wave that carries information or data through physical medium. Here the data is converted
into electromagnetic signal either as analog or digital and sent from sender to receiver.
Representation of signals:
Voltage and current are few time varying quantities that are used to represent data.
By varying these quantities with respect to time or frequency data can be transmitted.

Analog signal
Mathematical representation:
DC – Step input (sudden displacement)
x(t) = A u(t)

Ramp Input - (constant velocity)

x(t) = At u(t)

Parabolic input – (constant Acceleration)

𝐴𝑡 2
𝑥(𝑡) = 𝑢(𝑡)
2

Exercise:
(i) Draw a ramp signal of slope 5 units
X(t) = 5t u(t)

(ii) Draw a step signal of amplitude 4 unit

X(t) =4 u(t)
Unit impulse signal:

Sinusoidal signal
X(t) = A sin t u(t)
A – Amplitude
 = 2f rad/sec ; f – frequency in Hz.

Example:1
Sinusoidal signal with 50 Hz; f=50 Hz, T=1/f = 20msec.
X(t) = sin (2 x 50 x t) = sin 314 t

Example:2
Sinusoidal signal T=5msec. f = 1/T = 200 Hz
X(t) = 2 sin (2 x 200 x t) = sin 1256 t
 Sinusoidal signal with dealy in time

V1= 5 sin (2 x 0.5 x t) = 5 sin 3.14 t

V2 = 5 sin [(2 x 0.5 x t) - 45) = 5 sin (3.14 t - /4)

Exponential signals:
x(t) = a e bt
Case 1: a is real and positive and b is real and positive : 2 e2t u(t)
Case 2: a is real and positive and b is real and negative: 2 e-2t u(t)
Case 3: a is real and positive and b is complex with positive real : 2 e(2+j3)t u(t)
Case 4: a is real and positive and b is complex with negative real: 2 e-(2+j3)t u(t)
Graphical form of representation:
Case 1: x(t) = 2 e-2t u(t) - Exponential decay signal

Case 2: x(t) = 2 e2t u(t) - Exponential raising signal

 Case 3 : x(t) = 2 e2t sin 3t u(t) - Exponentially raising sinusoidal signal

Case 4: x(t) = 2 e-2t sin 3t u(t) - Exponentially decaying sinusoidal signal

Periodicity of Continuous Signal :

A continuous signal x(t) is periodic if it satisfies the condition
x(t) = x(t+T) where T is the time period of the signal
Example 1:
x(t) = sin 3t
f = 3/2 ; T =2/3
x(t+T) = sin 3(t+ 2/3) = sin (3t+2) = sin 3t =x(t)
Hence the signal is periodical
Example 2:
x(t) = e 2t
x(t+T) = e (2t+T) = e 2t e T  x(t)
Hence the signal is not periodical
Example 3:
X(t) = sin 3t + cos (2 /3) t
Two signals are added
T1 = 2/3 ; T2 = 3
𝑇1 2𝜋/3 2𝜋
= = is not a rational
𝑇2 3 9

Hence the composite signal is not periodic

Example 4:
X(t) = sin ( /8) t + cos (2 /3) t
Two signals are added
T1 = 16 ; T2 = 3
𝑇1 16
= is rational
𝑇2 3

Hence the composite signal is periodic

Resultant Time period T is LCM (T1,T2) = LCM(16,3) = 48 Sec
Example 5:
x(t) = sin2 3t
x(t) = (1+cos 6t) / 2 (2sin2t = 1+cos 2t)
1 cos 6𝑡
x(t) = 2 + ; The signal is addition of a DC signal and sinusoidal signal.
2

Hence it is periodic
Time period is T =  /3 sec
Analog to Digital Conversion:
The conversion of a physical analog signal to a discrete set of finite-precision numbers is
called signal sampling. It includes the following steps.
1. Sample-and-hold: An analog signal x(t) is sampled to a discrete-time signal x(nT).

2. Quantization: A discrete-time signal x(n) is converted to a quantized signal xq(n).

Amplitude of the signal is discrete and round off
 3. Encoding: A quantized discrete-time signal xq(n) is converted to a digital code c(n).

Analog to Digital Conversion process:

Sampling theorem:
Sampling theorem guides for the question: How many samples are necessary for preserving the
information contained in the signal?
To preserve the full information in the signal, it is necessary to sample at twice the maximum frequency
of the signal. This is known as the Nyquist rate.
Shannon’s Sampling Theorem:
A signal can be exactly reproduced if it is sampled at a frequency F, where F is greater than twice the
maximum frequency in the signal.
Fs  2 Fmax
Alaising:
If the signal is sampled at a frequency that is lower that the Nyquist rate, during reconstruction
from digital to analog (continuous time signal), aliasing will result. Aliasing is the presence of unwanted
components in the reconstructed signal. These components were not present when the original signal was
sampled. In addition, some of the frequencies in the original signal may be lost in the reconstructed
signal.
Discrete signalrepresentation:
𝑥(𝑛) = 𝑥(𝑡)|𝑡→𝑛𝑇
Where T is the sampling time. Sampling frequency is fs = 1/T
Note: Generally T is suppressed in the representation.
Discrete Step signal:
Unit step

Discrete Unit ramp

Discrete Unit parabolic signal

Discrete Impulse signal

Discrete Exponential : x[n] =C an u(n)
x(t) = C e-bt |t→nT
x(nT) = C e-b nT = C an where a = e-bT
x[n] = C an u(n)
Case 1: a > 1 : Exponential raising signal
Example : x[n] = 3n u(n)

Case 2: 0 < a < 1 : Exponential decay signal

Example : x[n] = (0.3)n u(n)

Case 3: -1 < a <0 : alternating and decaying Exponential signal

Example : x[n] = (-0.3)n u(n)

Case 4: -1 < a : alternating and raising exponential signal

Example : x[n] = (-3)n u(n)
Discrete time sinusoidal signal
x(t) = A sint | t→ n Ts
x(nTs) = A sin( nTs ) = A sin(2 f nTs )
Ts
= 𝐴 𝑠𝑖𝑛 (2 n 𝑇 ) = A sin  n
Ts
Where  =2 rad
𝑇

Condition : A discrete sinusoid is periodic only if the frequency  should be a rational multiple of 2.

(a) x[n] = ej2n

= ejn
 =2 and it is not a rational multiple of 2.
Hence not periodic
(b) x[n] = cos (3/ 4) n
3
 =3/ 4 = 2 x 3/8 == 2 8

Time period is N =8 samples

(c) x[n] = cos (3/4) n +sin (5/7) n

3
1=3/ 4 = 2 x 3/8 == 2 8; N1= 8 Samples
5
2=5/ 7= 2 x 5/14 == 2 14 ; N2 = 14 samples

LCM (N1,N2) = LCM(8,14) = 56 Samples

Even and Odd signals
Even Signal : x(t) = x(-t) Odd signal x(t) = -x(-t)
 Even Signal : x(t) = x(-t) Odd signal x(t) = -x(t)
Cosine signal: x(t) = cos t sine signal : x(t) = sin t

Any signal can be written as sum of even part and odd signal of the signal.
x(t) = xe(t)+xo(t)
To find even part of the given signal x(t)
𝑥(𝑡)+𝑥(−𝑡)
xe(t) = 2

To find odd part of the given signal x(t)

𝑥(𝑡)−𝑥(−𝑡)
xo(t) = 2

Example :1
x(t) = x3+4x2+5x+9
xe(t) = 4x2+9
xo(t) = x3+5x
Example 2:
Find the even and odd parts of the given signal

Solution:

The even part is : The odd part is :

x(t)+x(-t)/2 x(t)-x(-t)/2
Example 2:
Find the even and odd parts of the given signal

Solution:

The even part is : The odd part is :

Power and Energy Signal - Continuous Time signal

Energy :
𝑇
𝐸 = lim ∫ |𝑥(𝑡)|2 𝑑𝑡
𝑇→∞ −𝑇

Power : (ability)
1 𝑇
𝑃 = lim ∫ |𝑥(𝑡)|2 𝑑𝑡
𝑇→∞ 2𝑇 −𝑇

A signal with finite energy is called an energy signal .

0 < E < ∞ (P = 0)
A signal with infinite energy and finite average signal power is called a power signal .
 0 < P < ∞ (E = ∞)

Example 1: Exponential decay signal - Aperiodic

x(t) = e-2t u(t)
Since u(t) is attached, signal starts from 0 to T

𝑇 ∞ ∞
1 1
𝐸 = lim ∫ |𝑥(𝑡)|2 𝑑𝑡 = ∫ |𝑒 −2𝑡 |2 𝑑𝑡 = ∫ 𝑒 −4𝑡 𝑑𝑡 = (𝑒 −∞ − 𝑒 0 ) = 𝐽𝑜𝑢𝑙𝑒𝑠
𝑇→∞ 0 0 0 −4 4

1 𝑇 1 𝑇 −2𝑡 2 1 1
𝑃 = lim ∫ |𝑥(𝑡)| 𝑑𝑡 = lim ∫ |𝑒 | 𝑑𝑡 = lim ( (𝑒 −𝑇 − 𝑒 0 )) = 0 𝑤𝑎𝑡𝑡𝑠
2
𝑇→∞ 2𝑇 −𝑇 𝑇→∞ 𝑇 0 𝑇→∞ 𝑇 −4

The Exponential decay signal is an Energy Signal

Example 2:
x(t) = Ae-jt : Exponential form Aej : Absolute value or |𝐴𝑒 −𝑗𝜔𝑡 | is A
𝑇 𝑇 𝑇
2
𝐸 = lim ∫ |𝑥(𝑡)|2 𝑑𝑡 = lim ∫ |𝐴𝑒 −𝑗𝜔𝑡 | 𝑑𝑡 = lim ∫ 𝐴2 𝑑𝑡 = lim 𝐴2 (2𝑇) = ∞
𝑇→∞ −𝑇 𝑇→∞ −𝑇 𝑇→∞ −𝑇 𝑇→∞

1 𝑇 1 𝑇 2 1 2
𝑃 = lim ∫ |𝑥(𝑡)|2 𝑑𝑡 = lim ∫ |𝐴𝑒 −𝑗𝜔𝑡 | 𝑑𝑡 = lim 𝐴 (2𝑇) = 𝐴2 𝑤𝑎𝑡𝑡𝑠
𝑇→∞ 2𝑇 −𝑇 𝑇→∞ 2𝑇 −𝑇 𝑇→∞ 2𝑇

The given trigonometric signal is exponential form is a Power signal

Example 3:
x(t) = A sin t : periodical sinusoidal signal
𝑇 𝑇
𝐸 = lim ∫ |𝑥(𝑡)|2 𝑑𝑡 = lim ∫ |A sin 𝜔𝑡|2 𝑑𝑡
𝑇→∞ −𝑇 𝑇→∞ −𝑇

𝑇 (1−cos 2𝜔𝑡) 𝐴2 𝑇 𝑇
= lim ∫−𝑇 𝐴2 𝑑𝑡 = lim [∫−𝑇 1 𝑑𝑡 − ∫−𝑇 cos 2 𝜔𝑡 𝑑𝑡 ]
𝑇→∞ 2 𝑇→∞ 2
2𝜋 𝑇
𝐴2 sin 2𝜔𝑡 𝑇 𝐴2 sin 2
𝑇
𝑡 𝐴2
= lim [2𝑇 − ( ) ] = lim [2𝑇 − ( 2𝜋 ) ] = lim 2𝑇 = ∞
𝑇→∞ 2 2𝜔 −𝑇 𝑇→∞ 2 2
𝑇
𝑇→∞ 2
−𝑇

1 𝑇 1 𝑇 1 𝐴2 𝐴2
𝑃 = lim ∫ |𝑥(𝑡)|2 𝑑𝑡 = lim ∫ |A sin 𝜔𝑡|2 𝑑𝑡 = lim 2𝑇 = 𝑤𝑎𝑡𝑡𝑠
𝑇→∞ 2𝑇 −𝑇 𝑇→∞ 2𝑇 −𝑇 𝑇→∞ 2𝑇 2 2
The given sinusoidal signal is a Power signal

Note: some signals are neither power nor energy types because they have unbounded power and energy.
For these,
E = ∞ and P = ∞
Exercise 1: x(t) = A u(t) ; step signal is a power signal
 𝐴
It is a power signal with P = 2 watts and E =0 J

Exercise 2: x(t) = At u(t) : Ramp signal is neither power nor energy signal
E = ∞ and P = ∞
Power and Energy Signal - Discrete Time signal
E = ∑𝑛=∞
𝑛=−∞|𝑥(𝑛)|
2

1 1
P = lim (2𝑁+1 ∑𝑛=𝑁 2
𝑛=−𝑁|𝑥(𝑛)| ) ∑𝑛=∞ 𝑛
𝑛=0 𝑎 = a<1
𝑁→∞ 1−𝑎

𝑁 𝑎=1
∑𝑛=𝑁−1
𝑛=0 𝑎 𝑛
= { 1−𝑎 𝑁
𝑎≠1
1−𝑎

Example 1: x(n) = 0.2 n u(n)

𝑛=∞
E = ∑𝑛=∞ 2 𝑛 2
𝑛=−∞|𝑥(𝑛)| = E = ∑𝑛=0 |0.2 | = ∑𝑛=∞
𝑛=0 |0.2|
2𝑛

1 1
= ∑𝑛=∞ 𝑛
𝑛=0 |0.04| = 1−0.04 = 0.96
As N→∞, it is zero
1 1 1 1−0.04𝑁
P = lim (2𝑁+1 ∑𝑛=𝑁 2
𝑛=−𝑁|𝑥(𝑛)| ) = lim (𝑁+1 ∑𝑛=𝑁 𝑛 2
𝑛=0 |0.2 | ) = lim (𝑁+1 ) =0
𝑁→∞ 𝑁→∞ 𝑁→∞ 0.96

Energy is finite and power is zero, hence energy signal

Example 1: x(n) = sin  n
𝑒 jΩ𝑛 −𝑒 −jΩ𝑛
x(n) = 2𝑗

For x(n) = e-jn

2
E = ∑𝑛=∞ 2 𝑛=∞
𝑛=−∞|𝑥(𝑛)| = E = ∑𝑛=−∞|𝑒
jΩ𝑛
| = ∑𝑛=∞
𝑛=−∞ 1
2
=∞
1 1 1
P = lim (2𝑁+1 ∑𝑛=𝑁 2 𝑛=𝑁
𝑛=−𝑁|𝑥(𝑛)| ) = lim (2𝑁+1 ∑𝑛=−𝑁 1) = lim (2𝑁+1 2𝑁 + 1) = 1 Watts
𝑁→∞ 𝑁→∞ 𝑁→∞

Given signal is power signal