Scribd
Upload
25 views

DSP PPT Iit

Uploaded by

sri054
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
25 views

DSP PPT Iit

Uploaded by

sri054
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 16

Digital Filter Design

Impulse Invariant Transformation


• Impulse response of the digital filter is h(n)
• h(n) is obtained by uniformly sampling
impulse response ha(t) of the analog filter
i.e h(n)  ha (t ) t  nT  ha (nT )
• ‘T’ is the sampling period
Ex: 1
H a (s)  2 t
ha (t )  e u (t )
s2
2 nT
h(n)  ha (nT )  e u (nT )
1
H ( z)  2T 1
1 e z
Note: 1 1
 piT 1
s  pi 1 e z
• Consider the distinct pole case
N
N
Ai
H a (s)   ha (t )   Ai e pit u (t )
i 1 s  pi i 1

N
h(n)  ha (nT )   Ai e pi nT u (nT )
i 1

N
Ai
H ( z)   piT 1
i 1 1  e z
 aT 1
sa 1  e (cos bT ) z
  aT 1 2 aT 2
( s  a)  b
2 2
1  2e (cos bT ) z  e z

b e  aT (sin bT ) z 1

( s  a)  b
2 2
1  2e  aT (cos bT ) z 1  e 2 aT z 2

b (1) m 1 d m 1 1
 m 1  siT 1
( s  si ) m
!(m  1) dsi 1  e z
1 1 s
H a (s)    2
( s  1)( s  s  1) s  1 s  s  1
2

1 s
H a (s)  
s 1  1
2
 3
2

s   
 2   2 

 1  1  3 
 s    
1  2  2   2 
H a (s)   
s 1  1   3   3  
2 2
1
2
 3 

2

s       s      
 2   2   2   2  2  
 
Required Digital Filter analog filter specifications
specifications (K1 at Ωu , K2 at Ωr’)
u r 
r
(K1 at ω u, K2 at ω’ r) u 
Ts Ts

Determine prototype analog lowpass


filter specifications (K1 at Ω=1 , K2 at
Ω=Ωr)

Design prototype filter

Obtain the required digital Obtain the analog filter by using


filter by using Impulse analog low-pass to ______
Invariant Transformation transformation
Ex: Design a Butterworth filter using the impulse invariance method for the following specifications:
0.8≤|H(jω|≤1 when 0≤ ω ≤0.2π
|H(jω|≤0.2 when 0.6π ≤ ω ≤ π

ω ω


T (T=1)
0.2
u 
T

0.6
r 
T
K1 = -1.938dB at Ωu= Ω1 = 0.2π
K2 = -13.97dB at Ωr’ = Ω2 = 0.6 π
n  1.707  1
H 2 (s)  2
s  s 1
 c  0.2309

0.5262
H a (s)  H 2 (s) s s  2
c s  1.0259 s  0.5262
0.3015 z 1
H ( z) 
1  1.0433 z 1  0.35842 z 2
Problems:

(1) Using the impulse invariance technique,


design a low pass digital filter that is
equiripple in pass band and monotonic in the
stop band. The filter pass band edge is at
0.1π rad with a ripple of 2.5dB or less , and
the stopband edge at 0.2π rad with
attenuation of 40dB or more. Use T=1.
(2) Consider an analog system function
sa
H a (s) 
( s  a)2  b 2
Determine the digital filter from the analog filter
by means of impulse invariance. When would
it produce good results?
Relation between analog and digital filter
poles
• Analog pole at • Corresponding digital
pole zat e piT  e  j T
s  pi    j z  e T e jT  z e j
1. Analog filter poles in 1. Digital filter poles
LHP of s-plane inside the circle |z|=1
2. Analog filter poles in 2. Digital filter poles
RHP of s-plane outside the circle |z|=1
3. Analog filter poles on 3. Digital filter poles on
imaginary axis of s- the circle |z|=1
plane
  Ts


Ts

You might also like