19AIE204: Introduction to Communication Systems

Prepared by: Dr. Arpita Thakre – 7 Aug 2020

Recap
LTI System
Impulse Response of LIT system
Numericals
Spectral Domain representation of signals
Frequency Domain Representation of Periodic Continuous Time Signals (Fourier Series)
Frequency Domain Representation of Periodic Discrete Time Signals (Discrete Time Fourier Series)
Frequency Domain Representation of Aperiodic Continuous Time Signals (Fourier Transform)
Frequency Domain Representation of Aperiodic Discrete Time Signals (DTFT)
Why we represent time domain signal as a linear combination of complex exponentials

1
 Spectral Domain Representation of Signals
x(t) is a continuous time periodic signal with period T, i.e., x(t+T) = x(t)
The Fourier series representation of x(t) is given by:

𝑤0 = 2𝜋 𝑇

x(t) can be represented as a linear combination of harmonically related complex exponentials

Example

+ +
= +

2
 Fourier Series Representation of Continuous Time Periodic Signals

Q. Determine the Fourier series representation of the periodic square wave shown in Fig. 10

1
Fig. 10

3
 Fourier Series Representation of Continuous Time Periodic Signals

Q. Determine the Fourier series representation of the periodic square wave shown in Fig. 10

Fig. 10

2𝑇1 sin(𝑘𝑤0 𝑇) 𝑤0 𝑇 = 2𝜋
𝑎0 = 𝑎𝑘 = , 𝑘 ≠0
𝑇 𝑘𝜋

4
 Fourier Series Representation of Discrete Time Periodic Signals

x[n] is a discrete time signal with fundamental period x[n+N] = x[n]

Fourier series representation of x[n] is given by

𝑤0 = 2𝜋 𝑁

x[n] can be represented as a linear combination of harmonically related complex exponentials

5
 Fourier Series Representation of Discrete Time Periodic Signals
Q. Determine the Fourier series representation of the periodic discrete time signal shown in Fig. 11

Fig. 11

Plot of ak values
6
 Spectral Domain Representation of Aperiodic Signals

x(t) is a continuous time finite energy aperiodic signal.

Fourier Transform of x(t), written as X(jω) is given by

ω: angular frequency, unit is radian/sec

j = sqrt(-1)

An alternative set of equations:

∞
𝑋(𝑓) = 𝑥(𝑡)𝑒 −𝑗2𝜋𝑓𝑡 𝑑𝑡
−∞ f: frequency, unit is Hz
∞
𝑥(𝑡) = 𝑋(𝑓)𝑒 𝑗2𝜋𝑓𝑡 𝑑𝑓
−∞

7
 Spectral Domain Representation of Aperiodic Signals
∞
𝑋(𝑓) = 𝑥(𝑡)𝑒 −𝑗2𝜋𝑓𝑡 𝑑𝑡
−∞ f: frequency, unit is Hz
∞
j = sqrt(-1)
𝑥(𝑡) = 𝑋(𝑓)𝑒 𝑗2𝜋𝑓𝑡 𝑑𝑓
−∞

1
1
δ(t) ∞
𝑋(𝑓) = 𝛿 𝑡 𝑒 −𝑗2𝜋𝑓𝑡 𝑑𝑡 = 1
−∞
t=0 f

sgn(t)
1 −𝑗
𝑋(𝑓) =
𝜋𝑓
t
-1
8
 Spectral Domain Representation of Aperiodic Signals
∞
𝑋(𝑓) = 𝑥(𝑡)𝑒 −𝑗2𝜋𝑓𝑡 𝑑𝑡
−∞ f: frequency, unit is Hz
∞
j = sqrt(-1)
𝑥(𝑡) = 𝑋(𝑓)𝑒 𝑗2𝜋𝑓𝑡 𝑑𝑓 sin(𝜋𝑓)
𝑠𝑖𝑛𝑐 𝑓 =
−∞ 𝜋𝑓

u(t) = (1+sgn(t)/2)
1 1 𝑗
𝑋 𝑓 = 𝛿 𝑓 −
2 2𝜋𝑓
t
T

rect(t/T)
1
Tsinc(Tf) 2/T 3/T

0
-T/2 T/2 t
1/T
f
9
 Spectral Domain Representation of Aperiodic Signals
∞
𝑋(𝑓) = 𝑥(𝑡)𝑒 −𝑗2𝜋𝑓𝑡 𝑑𝑡
−∞ f: frequency, unit is Hz
∞
j = sqrt(-1)
𝑥(𝑡) = 𝑋(𝑓)𝑒 𝑗2𝜋𝑓𝑡 𝑑𝑓
−∞

sin(𝜋𝑓)
𝑠𝑖𝑛𝑐 𝑓 =
𝜋𝑓

(1/B)rect(f/B)
1/B
1

sinc(Bt)
2/B 3/B -B/2 B/2 f
0

1/B
t
10
11
12
13
14
15
16
17
18
19