EE-330 : Digital Signal Processing
Instructor : Dr Tauseef ur Rehman
Lab Engineer : Mr Zain ul Hassan
Credit Hours : 3-1 - Week No 4
Semester : Spring 2025
1
2
Core Concepts of DSP
Quick Recap
Discrete-time Fourier Transform
(DTFT)
3
Summary for Convergence:
Property Satisfied in the Time- Guaranteed results in the
domain Frequency-domain
Absolute Summability DTFT exists & converges uniformly to a
(e.g., All FIR Systems) continuous function of frequency.
Square Summable (finite energy) but no DTFT exists & only Mean-square
absolutely summable. convergence is guaranteed. The
resulting DTFT may be discontinuous
1 sin(o n)
e.g., x[n] = u[n − 1], (e.g., Ideal lowpass filter).
n n
Sequences which are not finite energy DTFT may not exist at all. In few cases,
(e.g., x[n]=1 or unit step sequence, or the DTFT may be expressed as sum of
periodic signals) impulse trains (generalized functions).
Sampling & Periodicity across Time
Time
and Frequency:
Frequency
Continuous & Aperiodic in Time: A Aperiodic & Continuous CTFT: C
Example: P 1 O
− at E N
e u (t ) a + j T
R
I I
Discrete & Aperiodic in Time: Periodic & Continuous DTFT: N
O
U
Example: D 1 O
I U
a n u[n] C 1 − ae − j S
Continuous & Periodic in Time: Aperiodic (in general) & Discrete
P (impulses): D
cos(ot ) E 1 1
( − o ) + ( + o )
I
R S
I
2 2 C
Discrete & Periodic in Time: O Periodic & Discrete: (Sum of impulse R
E
D trains) T
cos(o n) I
1 1 E
C ( − o ) + ( + o ), .
2 2
Sampling & Periodicity across Time
and Frequency:
Time Frequency
Continuous & Aperiodic in Time: Aperiodic & Continuous CTFT:
Example: 1
− at
e u (t ) a + j
C A
O P
Discrete & Aperiodic in Time: N Periodic & Continuous DTFT: E
Example: T
I 1 R
I
D a n u[n] N
U
1 − ae − j O P
I O D E
S Continuous & Periodic in Time: U Aperiodic (in general) & Discrete I R
C S (impulses):
R C I
E cos(ot ) 1 1
( − o ) + ( + o ) O
T 2 2 D
E I
Discrete & Periodic in Time: Periodic & Discrete: (Sum of impulse C
trains)
cos(o n) 1 1
( − o ) + ( + o ), .
2 2
Symmetry Properties (General):
Signal DTFT X( e𝑗𝜔 )
𝑥[n]
1 x*[n] X * (e− j )
x*[− n] X * (e j )
2
3 Re{x[n]} X e (e j )
4
jIm{x[n]} X o (e j )
5 xe [n] X R (e j ) = Re{X(e j )}
6 xo [n] jX I (e j ) = j Im{X(e j )}
Symmetry Properties for Real
Signal
sequences:
DTFT
𝑥[n] X( e𝑗𝜔 )
For any real sequence 7 X (e j ) = X * (e− j ) ( DTFT is Conj. Symm. )
x[n]
8 X R (e j ) = X R (e− j ) ( Real part is even )
( Imag. part is Odd )
9 X I (e j ) = − X I (e− j )
10 X (e j ) = X (e− j ) ( Mag. is even )
j − j ( Phase is odd )
11 X (e ) = −X (e )
12 xe [n] ( even part of x[n] ) X R (e j )
13 xo [n] ( odd part of x[n] ) jX I (e j )
Properties of DTFT:
Relationship in Time Relationship in Frequency
Linearity Property:
1 ax[ n] + by[n] aX (e j ) + bY (e j )
Shift in time: Phase-factor in freq:
2
𝑥 𝑛 − 𝑛𝑑 , 𝑛𝑑 ∈ ℤ X (e j ) e − jnd
Mult. by Exp.: jo n
Shift in freq:
3 x[n]e X (e j ( −o ) )
Time-reversal:
4 x[ − n] X(e − j ) (for x[n] real, X
*
(e j )
Multiplication by n: Differentiation in Freq dX (e j )
5 nx[ n ] j
d
Convolution in Time: Multiplication in freq:
6 x[ n] y[n] X (e j ) Y(e j )
Multiplication in Time: Conv: 1
7 x[ n] y[n]
2
−
X (e j )Y (e j ( − ) ) d
Parseval’s Theorem (Plancherel Theorem):
• Energy in time-domain is equal to the energy
in frequency-domain.
1
j 2
x[n] = X (e ) d
2
8
n =− 2 −
• Inner product is equal in both domains:
1
9
n =−
x[n] y [n] =
*
2
−
X (e j ) Y* (e j )d
Important DTFT Pairs
Z-Transform
Introduction & Analysis for Digital
Filter Design
12
13
Limitations of DTFT
Solution: Modify the Fourier Transform by providing an additional degree of freedom to handle
cases of divergent sequences. The result is z-transform
14
z-Transform
Motivation: Choose r such that the product h[n] 𝒓^ −𝒏 is
absolutely summable for any value of α
15
z-Transform : Choice of r
16
z-Transform : Definition
17
z-Transform : Example 1
18
Singularities of H(z)
2
19
Example 1
20
Pole-Zero Plots
Relationship with other transforms:
• Discrete counterpart to Laplace Transform :
Im{z}
S-plane Im{s} Z-plane z = re j
s = + j
j
Re{z}
Re{s}
CTFT on jΩ-axis, i.e., =0 (Unit circle, r=1)
DTFT on unit-circle, i.e., (r=1)
X ( j) = X (s) |s = j X (e j ) = X (z) |z =e j
(CTFT) (Laplace Transform) (DTFT) (Z-Transform)
X (s) = CTFT {x(t) e− t } X ( z ) = DTFT {x[n]r − n } 21
Convergence of Z-transform:
• Region of Convergence (ROC): The set of
values of ‘𝑧’ for which the Z-transform power
series converges.
• Z-transform of 𝑥[𝑛] converges when the DTFT
−𝑛 Im{z}
of 𝑥 𝑛 𝑟 does.
Re{z}
ROC
23
Right Sided Sequence : Monotonic
24
Right Sided Sequence : Alternating
25
Properties of Right Sided Sequence
26
Left Sided Sequence : Monotonic
27
Left Sided Sequence : Alternating
28
Properties of Left Sided Sequence
Properties of the ‘Region of
Convergence’ (ROC)
30 Properties of ROC (for rational Z-transforms):
1) ROC will have one of the following
forms:
Im{z} Im{z} Im{z}
rL
rL rR
Re{z} rR Re{z} Re{z}
ROC
ROC ROC
z rL 0 rR z rL 0 rR z
30
31 Properties of ROC (for rational Z-transforms):
2) The Fourier transform of 𝑥[𝑛]
converges absolutely iff the ROC of the
Z-transform of 𝑥[𝑛] includes the unit
circle.
Im{z} Im{z} Im{z}
Re{z} Re{z} Re{z}
z =1 z =1 z =1
Stable Stable Unstable
31
32 Properties of ROC (for rational Z-transforms):
3) The ROC cannot contain any poles.
Im{z} Im{z}
Re{z} Re{z}
Valid ROC Invalid ROC
32
33
Properties of ROC (for rational Z-
transforms):
• 4)For finite-duration sequences, the ROC
is the entire z-plane, except possibly @
z=0, and/or z=Infinity.
𝑥[𝑛] ROC Poles
A [ n] Entire z-
No poles
plane
[n − 1] z 0 z =0
[n + 2] z z=
[n − 1] + [n + 2] 0 z z = 0,
33
34
Properties of ROC (for rational Z-
transforms):
• 5)For right-sided sequences, the ROC
extends outward from the outermost finite
pole in 𝑋(𝑧) and (possibly including z= ∞.)
Im{z}
Outermost
pole Re{z}
34
35
Properties of ROC (for rational Z-
transforms):
6) For left-sided sequences, the ROC
extends inward from the innermost non-
zero pole in 𝑋(𝑧) and (possibly including
z= 0).
Im{z}
Re{z}
Innermost
pole
35
36
Properties of ROC (for rational Z-
transforms):
• 7) If 𝑥[𝑛] is a two-sided sequence, the
ROC will consist of a ring in the 𝑧-plane,
bounded on the interior and exterior by a
pole. No poles inside ROC.
Im{z}
Re{z}
36
37
Properties of ROC (for rational Z-
transforms):
• 8) The ROC must be a connected region:
Im{z}
Im{z}
Re{z}
ffffbb Re{z}
Invalid ROC Invalid ROC
37
38
39
Z-Transform from Pole-Zero Plot
40
Complex Poles & Zeros
41
Complex Poles & Zeros
Complex Poles & Zeros 42
Varying α and Fixed ωo
Complex Poles & Zeros 43
Varying α and Fixed ωo
Complex Poles & Zeros 44
Varying ωo and Fixed α
Complex Poles & Zeros 45
Varying ωo and Fixed α
46
Linear Phase FIR Systems
47
Linear Phase FIR Systems
48
Linear Phase FIR Systems
49
Linear Phase FIR Systems
50
Linear Phase FIR Systems
51
Linear Phase FIR Systems
HW: Review Section 3 & Example 4.3 - Holton
Inverse Z-Transform
53 Method 1: Inspection Method:
• Example: X ( z) =
1
, z 1/ 2
−1
• Find 𝑥[𝑛]=? 1 − (1/ 2) z
------------------------------------------------------
Recall, 1
a u[n]
n
, z a
1 − az −1
So,
n
1
x[n] = u[n]
2
53
54
Method 2: Partial Fraction
Expansion (PFE):
• Example: 1
X ( z) = −1 −1
, z 1/ 2
(1 − (1/ 4) z )(1 − (1/ 2) z )
• 𝑥[𝑛]=?
• Rewrite using PFE,
−1 2
X ( z) = −1
+ −1
, z 1/ 2
1 − (1/ 4) z 1 − (1/ 2) z
• Now use inspection method:
n n
• 1 1
x[n] = − u[n] + 2 u[n]
4 2
54
55 Partial Fraction Expansion Method:
σ𝑀
𝑘=0 𝑏𝑘 𝑧
−𝑘
• Given 𝑋 𝑧 = σ𝑁 , [𝑎0 , 𝑏0 , 𝑎𝑛 , 𝑏𝑚 ≠ 0]
𝑘=0 𝑎𝑘 𝑧 −𝑘
– Case 1: 0 < 𝑀 < 𝑁 [Proper Fractions]
• Zeros: 𝑀 + 𝑁 − 𝑀 [@ 𝑜𝑟𝑖𝑔𝑖𝑛]
• Poles: 𝑁
– Case 2: 𝑀 ≥ 𝑁 ≥ 0 [Improper Fractions]
• Zeros: 𝑀
• Poles: 𝑁 + 𝑀 − 𝑁 [@ 𝑜𝑟𝑖𝑔𝑖𝑛]
• No poles/zeros at Infinity
55
56
Partial Fraction Expansion (PFE)
Method:
σ𝑀 𝑏
𝑘=0 𝑘 𝑧 −𝑘
• Given 𝑋 𝑧 = σ𝑁 −𝑘
𝑘=0 𝑎 𝑘 𝑧
– Case 1: 0 < 𝑀 < 𝑁 [Proper Fractions]
𝑎0 ς𝑀 −1
𝑘=0(1−𝑐𝑘 𝑧 )
•𝑋 𝑧 =
𝑏0 ς𝑁 −1
𝑘=0(1−𝑑𝑘 𝑧 )
– Case 1a: PFE for Simple Poles:
𝐴𝑘
» 𝑋 𝑍 = σ𝑁
𝑘=0 1−𝑑𝑘 𝑧 −1
» 𝐴𝑘 = 𝑋 𝑍 1 − 𝑑𝑘 𝑧 −1
ȁ𝑧=𝑑𝑘
– Case 1b: PFE with a Pole of Order 𝑠:
𝐴𝑘 𝐶𝑚
» 𝑋 𝑍 = σ𝑁
𝑘=0,𝑘≠𝑖 + σ𝑠𝑚=1
1−𝑑𝑘 𝑧 −1 1−𝑑𝑖 𝑧 −1 𝑚
1 𝑑 𝑠−𝑚
» 𝐶𝑚 = 1 − 𝑑𝑖 𝑤 𝑠 𝑋 𝑤 −1
𝑠−𝑚 ! −𝑑𝑖 𝑠−𝑚 𝑑𝑤 𝑠−𝑚 𝑤=𝑑𝑖−1
by Eq. 3.47
56
57
Partial Fraction Expansion (PFE)
Method:
σ𝑀
𝑘=0 𝑏𝑘 𝑧
−𝑘
• Given 𝑋 𝑧 = σ𝑁 −𝑘
𝑘=0 𝑎𝑘 𝑧
– Case 2: 𝑀 ≥ 𝑁 [Improper Fractions]
• 𝑋 𝑧 = σ𝑀−𝑁
𝑘=0 𝐵𝑘 𝑧
−𝑘 + Proper Fraction
– Coefficients 𝐵𝑘 are obtained by long division.
– Example 3.10
HW: Review Examples 4.4 to 4.10 - Holton
57
58 Example 3.10
2
1+2𝑧 −1 +𝑧 −2 1+𝑧 −1
• 𝑋 𝑧 = 3 1 = 1 , 𝑧 > 1.
1− 𝑧 −1 + 𝑧 −2 1− 𝑧 −1 1−𝑧 −1
2 2 2
• As M = N = 2 and the poles are all 1st-order, so
𝐴1 𝐴2
• 𝑋 𝑧 = 𝐵0 + 1 +
1−2𝑧 −1 1−𝑧 −1
• The constant 𝐵0 can be found by long division:
−1+5𝑧 −1 −1+5𝑧 −1
• Thus, 𝑋 𝑧 = 2 + 3 1 =2 + 1
1−2𝑧 −1 +2𝑧 −2 1−2𝑧 −1 1−𝑧 −1
58
59 Example 3.10 (Contd.)
−1+5𝑧 −1 −1+5𝑧 −1
• 𝑋 𝑧 =2+ 3 1 =2 + 1
1−2𝑧 −1 +2𝑧 −2 1−2𝑧 −1 1−𝑧 −1
• Using PFE, we get
−9 8
• 𝑋 𝑧 =2+ 1 +
1− 𝑧 −1 1−𝑧 −1
2
• The impulse response 𝑥[𝑛] thus is ?
1 𝑛
• 𝑥 𝑛 = 2𝛿 𝑛 − 9 𝑢[𝑛] + 8𝑢[𝑛]
2
59
60 Method 3: Power Series Expansion:
• Example:
−1
• 𝑥[𝑛]=? X ( z ) = log(1 + az )
• ------------------------------------------------
• Using the Taylor series expansion, we can
n +1 n
(−1) a − n
X ( z) =
write:
z
n =1 n
• Comparing term by term with Z-transform
formula:
−1 𝑛+1 𝑎𝑛
• 𝑥𝑛 = 𝑢[𝑛 − 1]
𝑛 60
Z-Transform & LTI Systems
62 Properties of Z-Transform
Sequence Transform ROC
ax1[n] + bx2 [n] aX 1 (z) + bX 2 (z) Rx1 Rx2
x[n − no ] − no Possible addition or
z X ( z) Rx deletion of 0 or ∞
n zo Rx
z x[n]
0
X (z/ z o )
(P/Z scaled by z0)
dX (z)
nx[ n ] −z Rx
dz
x*[n] * *
X (z ) Rx
62
63 Properties of Z-Transform (Contd.)
Sequence Transform ROC
1 Contains Rx
Re{x[ n]} [ X (z) + X* (z* )] (Possible p/z cancellation)
2
1
Im{x[ n]} [ X (z) − X* (z* )] Contains Rx
2j (Possible p/z cancellation)
x*[− n] X * (1/ z* ) 1 / Rx
Poles/Zeroes move to 1/z*
x1[n] x2 [n] X1 ( z) X 2 ( z) Contains Rx1 Rx2
(Possible p/z cancellation)
HW: Review Examples 4.14 to 4.22 - Holton
63
64 System Function:
Z
h[n] H ( z )
• Output of LTI Systems given by:
Y ( z) = H ( z) X ( z)
• Causality: ℎ[𝑛] is causal iff 𝐻(𝑧) has an
ROC that extends from the largest pole
to infinity (and includes infinity).
• Stability: ROC must include the unit
circle
• Stability & Causality:
– All the poles must be inside the unit circle.
64
65
Find System Function(Response) from
Difference Equations:
• Consider linear constant coefficient
difference equations of the form:
– σ𝑁 𝑎
𝑘=0 𝑘 𝑦 𝑛 − 𝑘 = σ𝑀
𝑘=0 𝑏𝑘 𝑥[𝑛 − 𝑘]
• System Function 𝐻(𝑧) can be found as
follows:
– Take 𝑍-transform for each term:
• σ𝑁
𝑘=0 𝑎𝑘 𝑧 −𝑘 𝑌(𝑧) = σ𝑀 𝑏 𝑧 −𝑘 𝑋(𝑧)
𝑘=0 𝑘
– Find the ratio 𝑌 𝑧 /𝑋(𝑧)
𝑌 𝑧 σ𝑀
𝑘=0 𝑏𝑘 𝑧
−𝑘
•𝐻 𝑧 = = σ𝑁 −𝑘
𝑋 𝑧 𝑘=0 𝑎𝑘 𝑧
65
QUESTIONS & FEEDBACK
66
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 ).