Scribd
Upload
3 views11 pages

Module 3.5

The document discusses the Impulse Invariance Method for IIR filter design, detailing the mathematical transformations from analog to digital filters. It provides examples of designing filters, including a third-order Butterworth filter, and highlights the disadvantages of the method, particularly regarding aliasing issues in high pass filters. The content is structured around theoretical explanations and practical applications of filter design techniques.

Uploaded by

soujath048
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
3 views11 pages

Module 3.5

The document discusses the Impulse Invariance Method for IIR filter design, detailing the mathematical transformations from analog to digital filters. It provides examples of designing filters, including a third-order Butterworth filter, and highlights the disadvantages of the method, particularly regarding aliasing issues in high pass filters. The content is structured around theoretical explanations and practical applications of filter design techniques.

Uploaded by

soujath048
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 11

ECT303 DIGITAL SIGNAL

PROCESSING
MODULE 3-PART V
IIR Filter Design using Impulse Invariance
Technique

Ms. Neethu Radha Gopan, Asst. Prof., Dept. of ECE, RSET, Kakkanad
Impulse Invariance Method

𝑁
𝑐𝑘
➢ Let 𝐻 𝑠 = ෍
𝑠 − 𝑝𝑘
𝑘=1
Where 𝑝𝑘 are the poles of the analog filter & 𝑐𝑘 are the coefficients in the partial fraction
expansion.
➢ The corresponding z transform using impulse invariant technique is given by
𝑁
𝑐𝑘
H(z) = ෍
1 − 𝑒 𝑝𝑘 𝑇 𝑧 −1
𝑘=1

➢ For high sampling rates ( for small T), the digital filter gain is high. Then we can use
𝑁
𝑇𝑐𝑘
𝐻(𝑧) = ෍
1 − 𝑒 𝑝𝑘 𝑇 𝑧 −1
𝑘=1
Q. For the analog transfer function, 𝐻 𝑠 = 2/(𝑠 2 + 3𝑠 + 2) ,determine H(z)
using Impulse Invariance Method. Assume T=1 sec.

2 2
Soln: Given, 𝐻 𝑠 = 2 =
𝑠 + 3𝑠 + 2 (𝑠 + 1)(𝑠 + 3)
2 𝐴 𝐵
Using partial fraction, 𝐻 𝑠 = = +
(𝑠 + 1)(𝑠 + 3) 𝑠 + 1 𝑠 + 2
Solving for A and B, 𝐴 = 2, 𝐵 = −2
2 −2
𝐻 𝑠 = +
𝑠 − (−1) 𝑠 − (−2)
2 −2
𝐻 𝑧 = −𝑇 −1
+
1−𝑒 𝑧 1 − 𝑒 −2𝑇 𝑧 −1
2 −2
For T=1sec 𝐻 𝑧 = −1 −1 +
1−𝑒 𝑧 1 − 𝑒 −2 𝑧 −1
2 −2
𝐻 𝑧 = −1
+
1 − 0.3678𝑧 1 − .1353𝑧 −1

0.465𝑧 −1
𝐻 𝑧 =
1 − .503𝑧 −1 + .04976𝑧 −2
Q) Design a third order Butterworth digital filter using impulse invariance
technique. Assume sampling period of T= 1 sec.

Soln: For N=3, The transfer function of a normalized Butterworth filter is given by:
1 𝐴 𝐵 𝐶
𝐻 𝑠 = = + +
(𝑠 + 1)(𝑠 2 + 𝑠 + 1) (𝑠 + 1) (𝑠 + 0.5 + 0.866𝑗) (𝑠 + 0.5 − 0.866𝑗)

Solving for A, B and C by partial fraction,

𝐴 = 1, 𝐵 = −0.5 + 0.288𝑗 , 𝐶 = −0.5 − 0.288𝑗


1 −0.5 + 0.288𝑗 −0.5 − 0.288𝑗
𝐻 𝑠 = + +
𝑠 − (−1) 𝑠 − (−0.5 − 0.866𝑗) 𝑠 − (−0.5 + 0.866𝑗)

Using Impulse Invariance Method,


1 −0.5 + 0.288𝑗 −0.5 − 0.288𝑗
𝐻 𝑧 = −1 −1 + (−0.5−0.866𝑗) −1
+
1−𝑒 𝑧 1−𝑒 𝑧 1 − 𝑒 (−0.5+0.866𝑗) 𝑧 −1
Q) Design a Butterworth digital filter using impulse invariance method for the following
specifications. Assume sampling period of T= 1 sec.
0.8 ≤ | H(𝑒 𝑗𝜔 ) | ≤ 1, for 0 ≤ 𝜔 ≤ 0.2𝜋 & | H(𝑒 𝑗𝜔 ) | ≤ 0.2, for 0.6 𝜋 ≤ 𝜔 ≤ 𝜋

1 1
Soln: Given 𝜔𝑝 = 0.2𝜋 𝑟𝑎𝑑, 𝜔𝑠 = 0.6𝜋 𝑟𝑎𝑑, = 0.8 , = 0.2
1+𝜀 2 1+𝜆2
1
= 0.8 → 𝜀 = 0.75
1+ 𝜀2
1
= 0.2 → 𝜆 = 4.899
1+ 𝜆2
To design the analog counterpart we require Ω𝑝 & Ω𝑠 . Using impulse invariance
transformation, 𝜔𝑝 = Ω𝑝 𝑇 & 𝜔𝑠 = Ω𝑠 𝑇
𝜔𝑝 𝜔𝑠
Ω𝑝 = = 0.2𝜋 rad/sec & Ω𝑠 = = 0.6𝜋 rad/sec
𝑇 𝑇

𝜆 4.899
log log
𝑁= 𝜀 = 0.75 = 1.71
Ω
Choose N=2
log 𝑠 log 3
Ω𝑝
𝑠𝑘 = Ω𝑐 𝑒 𝑗𝜙𝑘 , k =1,2,…. N (Take Ω𝑐 = 1𝑟𝑎𝑑/sec)
𝜋 2𝑘 − 1 𝜋
𝜙𝑘 = +
2 2𝑁
𝜋 𝜋 3𝜋
k=1, 𝜙1 = + = , 𝑠1 = 𝑒 𝑗3𝜋/4 = −0.707 + 𝑗0.707
2 4 4
𝜋 3𝜋 5𝜋
k=2, 𝜙2 = + = , 𝑠2 = 𝑒 𝑗5𝜋/4 = −0.707 − 𝑗0.707
2 4 4
1
The normalized butterworth filter is given by 𝐻 𝑠 =
(𝑠 + 0.707 − 𝑗0.707)(𝑠 + 0.707 + 𝑗0.707)
1
𝐻 𝑠 = 2
𝑠 + 1.414𝑠 + 1
Ω𝑝 0.2𝜋
Ω𝑐 = = = 0.231𝜋
𝜀 1/𝑁 (0.75 )1/2
𝑠 𝑠
The transfer function for the lowpass filter is obtained by substituting s→ = in H(s).
Ω𝑐 0.231𝜋
1
∴𝐻 𝑠 = 𝑠 𝑠
( + 0.707 − 𝑗0.707)( + 0.707 + 𝑗0.707)
0.725 0.725
0.526 𝐴 𝐵
𝐻 𝑠 = = +
(𝑠 + 0.51 + 0.51𝑗)(𝑠 + 0.51 − 0.51𝑗) 𝑠 + 0.51 + 0.51𝑗 𝑠 + 0.51 − 0.51𝑗

Solving for A & B , we get A= 0.516j & B= - 0.516j

0.516𝑗 0.516𝑗
∴𝐻 𝑠 = −
𝑠 − (−0.51 − 0.51𝑗) 𝑠 − (−0.51 + 0.51𝑗)
0.516𝑗 0.516𝑗
Using Impulse Invariance, H(z) = −0.51−0.51𝑗 𝑇 𝑧 −1
− −0.51+0.51𝑗 𝑇 𝑧 −1
1−𝑒 1−𝑒
Given T =1 sec

0.516𝑗 0.516𝑗 .3019𝑧 −1


∴ 𝐻(𝑧) = − =
1−𝑒 −0.51−0.51𝑗 𝑧 −1 1−𝑒 −0.51+0.51𝑗 𝑧 −1 1 − 1.048𝑧 −1 + 0.36𝑧 −2
Disadvantage of Impulse Invariance Technique

➢ The relation between s and z is given by z = 𝑒 𝑠𝑇


➢ Consider 2 different poles 𝑠1 and 𝑠2
𝑠1 = 𝜎 + 𝑗Ω
2𝜋
𝑠2 = 𝜎 + 𝑗 Ω +
𝑇
(𝜎+𝑗Ω)𝑇 = 𝑒 𝜎𝑇 . 𝑒 𝑗Ω𝑇
𝑧1 = 𝑒 𝑠1𝑇 = 𝑒
2𝜋
𝜎+𝑗 Ω+ 𝑇 𝑇
𝑠2 𝑇
𝑧2 = 𝑒 = 𝑒
= 𝑒 𝜎𝑇 .𝑒 𝑗Ω𝑇 . 𝑒 𝑗2𝜋 = 𝑒 𝜎𝑇 . 𝑒 𝑗Ω𝑇
➢ Both these poles 𝑠1 & 𝑠2 map to the same location (𝑧1 = 𝑧2 ) in the z-plane.
➢ There are infinite number of poles in the s-plane that map to the same location in the z-
plane.
➢ These poles have the same real part and their imaginary parts differ by some integer
2𝜋
multiple of .
𝑇
➢ Due to the presence of this aliasing, the impulse invariant method is unsuccessful for the
implementation of digital high pass filters.
11 END of PART -V

THANK YOU!

You might also like