EE-330 : Digital Signal Processing

Instructor : Dr Tauseef ur Rehman

Lab Engineer : Mr Zain ul Hassan
Credit Hours : 3-1 - Week No 3
Semester : Spring 2025

1
 2

Core Concepts of DSP

Impulse Response

3
 4

Impulse Response
 Types of Systems
1 Finite Impulse Response (FIR) – Systems with impulse
responses that are finite in time duration

RECALL – Moving-Average Example of Discrete Systems

 Discrete-time System
#5 : Moving-Window Average
FIR Systems : Moving-Average
 Types of Systems
2 Infinite Impulse Response (IIR) – Systems with impulse
responses that are infinite in time duration

RECALL – LCCDE Example of Discrete Systems

 Discrete-time System
#7 : Linear Const-Coeff Diff Eqn (LCCDE)
IIR Systems : LCCDE Ex 1
Convolution

11
Convolution
Convolution (Sum / Integral)
Convolution is Commutative
Convolution is Distributive
Convolution is Associative
LTI System : Causality & Memoryless
LTI System : Invertibility
Convolution : Direct-Summation
Convolution : Direct-Summation
Convolution : Direct-Summation
Convolution : Flip & Shift
Convolution : Flip & Shift
Convolution : Flip & Shift
Convolution : Flip & Shift
Convolution : Flip & Shift
Convolution : Flip & Shift
Convolution : Flip & Shift
Convolution : Flip & Shift
Convolution : MATLAB
Convolution : Matrix Form
Convolution : Matrix Form
 Convolution : Implementation Issues

• In real time systems input maybe available point by

point or frame by frame
• Direct-Summation or Matrix Multiplication methods
are not very practical
• Sufficient Memory
• Minimum number of Multiplications and Addition
operations
• Minimum number of storage elements and data
moves
• DSP Processors and Custom FPGA implementations
34
Digital Filters

35
Digital Filters
Digital Filters
Digital Filters
Digital Filters
Digital Filters
Digital Filters
Digital Filters
 Eigen-functions for LTI Systems:

• Complex exponentials are eigenfunctions of LTI systems:

xn = e jn
• Let’s see what happens if we feed x[n] into an LTI system:
 
yn =  hkxn − k =  hke j(n −k )

k = − k = −

• The eigenvalue is called the frequency response of the

system   − jk  jn
yn =   hk e j jn
e = He e ( ) Eigenfunction
 k = − 
• H e( ) is a complex function of frequency
j Eigenvalue

( ) =  hke

j − jk
He
k = −
43
 Frequency Response:
j
• H (e ) is called the frequency-response of
the LTI system.
• It is complex-valued and can be expressed as
j j j
H (e ) = H R (e ) + j H I (e )
(Real & Imaginary part representation)

j j H (e j )
H (e ) = H (e ) e
(Magnitude & phase representation)

44
 Example: Frequency Response of the Ideal Delay

• LCCDE Representation: y[n] = x[n − n d ]

x[n] = e jn
• For input
• The output will be
j (n − nd ) j n − j nd
y[n] = e =e e
• The eigen-value is j
H (e ) = e − j nd

H R (e j ) = cos( n d ) H (e j ) = 1
j
H I (e ) = − sin( n d ) H (e j ) = − n d

(Real & Imaginary part representation) (Magnitude & phase representation)45

Example : Sinusoidal Response of LTI Systems

• For input x[n] = A cos(o n +  )

A j jo n A − j − jo n
x[n] = e e + e e
2 2 (Inverse Euler’s Formula)
A j jo n A j
e e jw
e H (e jo n ) e jo n
2 H (e )
2
A − j − jo n A − j
e e jw
e H (e − jo n ) e − jo n
2 H (e )
2

• The general output will be

A
y[ n] = [H(e jo ) e j e jo n + H(e − jo ) e − j e − jo n ]
2 46
 Example (Contd)
• Assuming an ideal-delay system
j − j nd
H (e ) = e

• The output will be

A − jo nd j jo n
y[ n] = [e e e + e + jo nd e − j e − jo n ]
2
y[n] = A cos(o n +  − o n d )

y[n] = A cos[o (n − n d ) +  ]
Note: Output is the input delayed by nd
47
 Periodicity of the Frequency Response:

• Frequency-response of discrete-time LTI

systems is always periodic with period 2𝜋

• It can be shown from definition and periodicity

of the complex exponential:
– 𝐻 𝑒 𝑗(𝜔+2𝜋𝑟) = 𝐻 𝑒 𝑗𝜔 , ∀𝑟 ∈ ℤ.

48
Examples of Frequency Responses:
1) Ideal Lowpass Filter
H lp (e j )
One period of the frequency response
1 Fundamental Period

− −c 0 c  

H lp (e j )
Periodicity of the frequency response
1

−2 −2 + c − −c 0 c  2 − c 2 

49
Examples of Frequency Responses:
2) Ideal Highpass Filter

H hp (e j ) Fundamental Period

− −c 0 c  

50
Examples of Frequency Responses:
3) Ideal Bandpass Filter

H bp (e j ) Fundamental Period

− −b −a 0 a b  

51
Examples of Frequency Responses:
4) Ideal Bandstop Filter

Hbs (e j ) Fundamental Period

− −b −a 0 a b  

52
Understanding Frequency Domain
Representation of Discrete-Time
Signals & Systems
Discrete-time Fourier Transform
(DTFT)

53
 Lesson Objectives
1. Understand the definition and mathematical
formulation of DTFT.
2. Learn the conditions for the existence of DTFT.
3. Explore the properties of DTFT.
4. Analyze DTFT of finite and infinite sequences
5. Apply DTFT to discrete-time systems.
 Introduction to DTFT
• DTFT is a tool for analysing discrete-time signals
in the frequency domain.
• It extends the concept of the Continuous-Time
Fourier Transform (CTFT) to discrete-time signals.
• Widely used in digital signal processing,
communications, and system analysis.
 Recap: CTFT
∞
• CTFT: X jΩ = ‫׬‬−∞ x t e−jΩt dt

• Converts continuous-time signals to the

frequency domain

• Limitations: Cannot directly analyse discrete-

time signals.
 Recap: Discrete-Time Signals
• Defined as sequences 𝑥 𝑛 , where n is an
integer.

• Examples: 𝑥 𝑛 = {1,2,3,4} or 𝑥 𝑛 = δ 𝑛

• Frequency domain analysis helps understand

signal behaviour.
 DTFT : Definition
• DTFT: 𝑋 𝑒 𝑗ω = σ∞
𝑛=−∞ 𝑥 𝑛 𝑒 −𝑗ω𝑛

• 𝑥 𝑛 : Discrete-time signal.
• ω : Normalized frequency (radians/sample).
• 𝑋 𝑒 𝑗ω : Frequency domain representation.
 DTFT : Definition
The DTFT of a discrete-time signal(𝑥 𝑛 ) is defined as:
∞

𝑋 𝑒 𝑗ω = ෍ 𝑥 𝑛 𝑒 −𝑗ω𝑛
𝑛=−∞
Derivation:
1. Start with the continuous-time Fourier transform (CTFT):
∞
𝑋 𝑗Ω = න 𝑥 𝑡 𝑒 −𝑗Ω𝑡 𝑑𝑡
−∞
2. For discrete-time signals, replace ( 𝑡 ) with ( 𝑛𝑇 ), where ( 𝑇 ) is
the sampling period:
∞

𝑋 𝑒 𝑗ω = ෍ 𝑥 𝑛 𝑒 −𝑗ω𝑛
𝑛=−∞
3. Here, ( ω = Ω𝑇 ) is the normalized frequency.
 DTFT : Frequency Domain
Representation
• ( ω) ranges from ( −π) 𝑡𝑜 ( π).

• DTFT is periodic with period ( 2π).

• Interpretation: (𝑋 𝑒 𝑗ω )describes the frequency content

of(𝑥 𝑛 ).
 DTFT : Existence
𝑇ℎ𝑒 𝐷𝑇𝐹𝑇 𝑒𝑥𝑖𝑠𝑡𝑠 𝑖𝑓 𝑡ℎ𝑒 𝑠𝑖𝑔𝑛𝑎𝑙 (𝑥 𝑛 ) 𝑖𝑠 𝑎𝑏𝑠𝑜𝑙𝑢𝑡𝑒𝑙𝑦 𝑠𝑢𝑚𝑚𝑎𝑏𝑙𝑒:
σ∞𝑛=−∞ 𝑥 𝑛 <\inf

𝐏𝐫𝐨𝐨𝐟:
The DTFT is defined as:
∞

𝑋 𝑒 𝑗ω = ෍ 𝑥 𝑛 𝑒 −𝑗ω𝑛
𝑛=−∞
𝐹𝑜𝑟 (𝑋 𝑒 𝑗ω ) 𝑡𝑜 𝑐𝑜𝑛𝑣𝑒𝑟𝑔𝑒, 𝑡ℎ𝑒 𝑚𝑎𝑔𝑛𝑖𝑡𝑢𝑑𝑒 𝑜𝑓 𝑡ℎ𝑒 𝑠𝑢𝑚 𝑚𝑢𝑠𝑡 𝑏𝑒 𝑓𝑖𝑛𝑖𝑡𝑒:
∞ ∞

𝑋 𝑒 𝑗ω ≤ ෍ 𝑥𝑛 𝑒 −𝑗ω𝑛 = ෍ 𝑥 𝑛
𝑛=−∞ 𝑛=−∞
𝑇ℎ𝑢𝑠, 𝑎𝑏𝑠𝑜𝑙𝑢𝑡𝑒 𝑠𝑢𝑚𝑚𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑒𝑛𝑠𝑢𝑟𝑒𝑠 𝑐𝑜𝑛𝑣𝑒𝑟𝑔𝑒𝑛𝑐𝑒.
 DTFT : Finite Length Sequences
Example: Rectangular pulse (𝑥 𝑛 = {1,1,1,1})𝑓𝑜𝑟(𝑛 = 0,1,2,3).

Derivation:

Apply the DTFT definition:

3

𝑋 𝑒 𝑗 𝜔 = ෍ 𝑥 𝑛 𝑒 −𝑗 𝜔𝑛
𝑛=0

Substitute (𝑥 𝑛 = 1):

𝑋 𝑒 𝑗 𝜔 = 1 + 𝑒 −𝑗 𝜔 + 𝑒 −𝑗 2𝜔 + 𝑒 −𝑗 3𝜔

Simplify using the geometric series formula:

1 − 𝑒 −𝑗 4𝜔
𝑋 𝑒𝑗 𝜔 =
1 − 𝑒 −𝑗 𝜔
 DTFT : Infinite Length Sequences
Example: Exponential sequence (𝑥 𝑛 = 𝑎𝑛 𝑢 𝑛 )𝑓𝑜𝑟( 𝑎 < 1).

*Derivation:*

1. Apply the DTFT definition:

∞

𝑋 𝑒 𝑗 𝜔 = ෍ 𝑎𝑛 𝑒 −𝑗 𝜔𝑛
𝑛=0

2. Combine terms:
∞
𝑛
𝑋 𝑒 𝑗 𝜔 = ෍ 𝑎𝑒 −𝑗 𝜔
𝑛=0

3. Use the geometric series formula:

𝑗𝜔
1
𝑋 𝑒 =
1 − 𝑎𝑒 −𝑗 𝜔
Important DTFT Pairs
 Important DTFT Pairs

HW: Plot the signals x[n] and their respective DTFTs for all above.
Properties of the DTFT
 DTFT Property: Symmetry
Time Reversal
 DTFT Property: Symmetry
Even – Odd Symmetry
 DTFT Property: Symmetry
Even – Odd Symmetry
 DTFT Property: Symmetry
Conjugate Symmetry & Asymmetry

1 2

2
 DTFT Property: Symmetry
Conjugate Symmetry & Asymmetry

Since,

Even
Odd
 DTFT Property: Symmetry
Conjugate Symmetry & Asymmetry

Since,

Even

Neither Even or Odd

 DTFT Property: Symmetry
Consequence of Symmetry
 DTFT Property: Symmetry
Important Symmetry Relations for DTFT
 DTFT Property: Symmetry
Consequence of Symmetry
 DTFT Property: Parseval’s Theorem
Parseval's theorem states:
∞
2
1 π 2
෍ 𝑥𝑛 = න 𝑋 𝑒 𝑗ω 𝑑
2π −π
𝑛=−∞
Proof:
1. Start with the energy of (𝑥 𝑛 ):
∞ ∞
2
෍ 𝑥𝑛 = ෍ 𝑥 𝑛 𝑥∗ 𝑛
𝑛=−∞ 𝑛=−∞
2. Express (𝑥 ∗ 𝑛 ) using the inverse DTFT:

𝑛
1 π 𝑒 𝑗ω
𝑥 = න 𝑋 𝑒 𝑗ω𝑛 𝑑
2π −π
 DTFT Property: Parseval’s Theorem
3. Substitute into the energy expression:
∞
1 π ∗ 𝑗ω 𝑗ω𝑛
෍ 𝑥𝑛 න 𝑋 𝑒 𝑒 𝑑ω
2π −π
𝑛=−∞
4. Swap summation and integration:

π ∞
1
න 𝑋 ∗ 𝑒 𝑗ω ෍ 𝑥 𝑛 𝑒 𝑗ω𝑛 𝑑
2π −π
𝑛=−∞
5. Recognize the DTFT of (𝑥 𝑛 ):
1 π ∗ 𝑗ω 1 π
2
න 𝑋 𝑒 𝑋 𝑒 𝑗ω 𝑑ω = න 𝑋 𝑒 𝑗ω 𝑑
2π −π 2π −π
QUESTIONS & FEEDBACK
78
 DTFT Property: Symmetry

-For real-valued signals:

- (𝑋 𝑒 𝑗ω ) is conjugate symmetric:
(𝑋 𝑒 𝑗ω = 𝑋 ∗ 𝑒 −𝑗ω ).

- Magnitude is even: ( 𝑋 𝑒 𝑗ω = 𝑋 𝑒 −𝑗ω ).

- Phase is odd: (∠𝑋 𝑒 𝑗ω = −∠𝑋 𝑒 −𝑗ω ).
 DTFT Property: Linearity

If (𝑥1 𝑛 𝑋1 𝑒 𝑗ω ) and (𝑥2 𝑛 𝑋2 𝑒 𝑗ω ),

then:
(𝑎𝑥1 𝑛 + 𝑏𝑥2 𝑛 𝑎𝑋1 𝑒 𝑗ω + 𝑏𝑋2 𝑒 𝑗ω ).

Example: (𝑥 𝑛 = 2δ 𝑛 + 3δ 𝑛 − 1 ).