Digital Signal Processing

(BEC 503)
Unit-2
IIR Digital Filter Design
Dr. Neelesh Kumar Gupta
Professor
Department of Electronics and Communication Engineering
Ajay Kumar Garg Engineering College, Ghaziabad
 Syllabus

Unit-2

Infinite Impulse Response Digital (IIR) Filter Design:

Introduction to Filters, Impulse Invariant Transformation, Bi-Linear
Transformation, All- Pole Analog Filters: Butterworth and Chebyshev,
Design of Digital Butterworth and Chebyshev Filters, Frequency
Transformations.
Unit-2

Infinite Impulse Response Digital (IIR) Filter Design:

Introduction to Filters
Impulse Invariant Transformation
Bi-Linear Transformation
All- Pole Analog Filters: Butterworth and Chebyshev
Design of Digital Butterworth and Chebyshev Filters
Frequency Transformations
 Introduction of Digital Filters
Filter means Frequency selective Electronic Networks.

The system implementing this operation is called a filter.

❑ The discrete-time system for the treatment of discrete-time signal is called a digital
filter.

❑ Digital filters are implemented using a digital computer or special purpose digital
hardware.
❑ These are programmable. Hence the filter coefficients can be varied at any time to
implement adaptive features in digital filters.
❑ In practice the magnitude response specifications of a digital filter in the pass band
and in the stop band are given with some acceptable tolerances.
Analog Vs Digital Filters

• Digital Filters process discrete signals with software & digital hardware while
analog process continues time signal with Lumped elements (R, L & C) & ICs.

• Digital filters are represented by difference equation and are flexible but analog
are presented in terms of system components with less flexibility.

• Interference, noise and environment effects are minimum in digital filters with
portability.
Classification of Digital Filters

On the basis of magnitude characteristics of the transfer function of filters:

Low pass filters

High pass filters

Band pass filters

Band stop filters.

Active & Passive Filters
(on the basis of components used)

• An active filter is one that, along with R, L, and C components, also contains an
energy source, such as that derived from an operational amplifier.

• A passive filter is one that contains only R, L, and C components without any energy
source. It is not necessary that all three be present.
On the basis of length of their impulse response sequences

Infinite impulse response (IIR) filters (R+ NR)

b r z −r

H ( z) = r =0
N
1+ a
k =1
k z −k
 Contd:

Finite impulse response (FIR) filters (NR)

N −1
H ( z ) =  h( n) z −n

n =0
Contd:
Contd:
 IIR FILTER DESIGNING METHODS

1. Approximation of derivatives

2. Impulse Invariant Transformation

3. Bi-Linear Transformation
 IIR Filter Design
The objective of digital filter design is to develop a transfer function H(z) meeting the
frequency response specifications.
• Let an analog transfer function be

Pa ( s )
H a (s) =
Da ( s )
where the subscript “a” indicates the analogue domain

• A digital transfer function derived from this is denoted as

H (Z) = N (Z)
---------
D (Z)
Contd:

Basic idea behind the conversion of H (S) into H(Z) is to apply a mapping from the s-
domain to the z-domain so that essential properties of the analogue frequency response
are preserved

Thus mapping function should be such that Imaginary axis in the s-plane be mapped
onto the unit circle of the z-plane

A stable analogue transfer function be mapped into a stable digital transfer

function
 Approximation of Derivatives Method
It is one of the simplest method for converting an analog filter into digital filter is
to approximate the differential equations by an equivalent difference equation.

Put S= 1-z -1
----- into H (S) and find H (Z) Here T is
T sampling interval in seconds.

This method is often used to solve a linear constant coefficient

differential equation numerically on digital computer.
Example

If H (S) = 1
---------------
(S+0.1) 2 +9

For finding H (z) Put S= 1 - Z -1 /T

It is rarely used due to complexity than analog filter is some

applications.
Most common methods of converting analog filters into equivalent digital filter

• Impulse invariant method-

It has equal impulse responses between equivalent digital filter and analog filter at
sampling point

• Bilinear transformation

Analog transfer function in s-plane is mapped into the discrete transfer function in
z-plane or jw axis in the s-plane is mapped onto the unit circle in Z-plane
Contd:

Designing Steps:

Analog Filter Digital Filter

Transfer Function Transfer Function
H (s) H ( z)

Impulse Response Impulse Response Series

h(t ) h( nT )
Derivation of Impulse Invariant Method

The transformation technique can be well understood by first considering a simple

distinct pole case for the analog filter’s system function, as shown below.
Contd:

The impulse response of the system specified can be obtained by taking the inverse Laplace
transform and it will be of the form

where ua(t) is the unit step function in continuous time.

The impulse response h(n) of the equivalent digital filter is obtained by uniformly sampling
ha(t)
Contd:
The system response of the digital system of eq., can be obtained by taking the
z-transform, i.e.

Interchanging the order of summation,

Contd:

• Now, by comparing Equations, the mapping formula for the impulse invariant
transformation is given by
Relation between analog and digital poles

• Therefore, the analog poles and the digital poles are related by the relation

• The general characteristic of the mapping can be obtained by substituting and

expressing the complex variable z in the polar form as
• With these substitution
Similarly we can derive remaining two formulas also given
below as:
Example 1:

If H (S) = 2 Find H (Z) for

----------------
S2 + 3S +2 T = 1 s & T = 0.1 s
Sol: H (S) = 2
-----------------
(S+1) (S+2) further by using partial fraction it can be

H (s)= 2/s+1 + -2/s+2

Determine H (Z) using linear factor formula by

putting value of T. H (z) = 0.465 z -1
-----------------

1-0.30 z -1 + 0.049 z -2
Example 2:

We can apply the below mentioned formulas for solving any problem

3
Contd:
Contd:
Contd:
Contd:

Advantages:

• Stable design.

• Analog frequency and digital frequency are linearly related.

Limitations:

• Impulse Invariant method can only be used to realize low pass filters and
a limited class of band-pass filters.
Many to one mapping Problem

We can’t design high pass filters or certain band-reject filters using this method.

• Moreover, the many to one mapping in the impulse invariance method

(s-domain to z-domain) causes the issue of aliasing, which is highly undesirable.

• The value of sampling time T can be chosen very small to mitigate the effect of aliasing
or another transform such as Bilinear transform removes that issue by using one-to-one
mapping.

To overcome these limitation a mapping from S plane to Z plane was to be developed is

called, Bi-Linear transformation Method.
Bilinear Transformation Method
• Mapping of s-plane into the z-plane (One to One Mapping)

This is a conformal mapping that transforms the imaginary axis of s plane into
the unit circle of z plane. Left side of s plane is transformed into inside the unit
circle of z plane and vice versa.
Contd:
Ω = 2/T tan ω/2
Contd:
Bilinear transformation
−1
2 1− z
1 + ( sT ) / 2 s =
z= T 1+ z −1
1 − ( ST ) / 2
 Derivation of Bi-Linear Transformation Method
• Also, the transformation of a stable analog filter results in a stable digital filter as all the
poles in the left half of the s-plane are mapped onto points inside the unit circle of the
z-domain.
• The bilinear transformation is obtained by using the trapezoidal formula for numerical
integration. Let the system function of the analog filter be:

• The differential equation describing the analog filter can be obtained from above Eqn
as:
Bilinear Transformation (cont.)

(1)

(2)

(3)
(4)
(5)

(6)
 (7)

(8)

From Eq. (7), it can be noted that if r < 1, then σ < 0, and if r > 1, then σ > 0.
• Thus, the left-half of the s-plane maps onto the points inside the unit circle in the z-plane and the
transformation results in a stable digital system.
• Consider Eq. (8), for unity magnitude (r = 1), σ is zero.
(9)

(10)
Contd:
So in Bilinear transformation:

2
2 1− z −1 s+
s=  z= − T
−1
T 1+ z 2
s−
T

2 p
p = tan( )
T 2
Example 1:
If H (S) = 2 Find H (Z) for
----------------
S2 + 3S +2 T = 1 s & T = 0.1 s
Sol: Put S = 2/T (1-Z -1 /1+Z -1)

For H (Z) using linear factor formula by putting value of T.

H (Z)= 0.1667 Z 2 + 0.33 Z 1 + 0.166

________________________
Z 2 - 0.33 Z 1
Example 2:
Example 3:

Convert the analog filter into digital filter The digital filter is to have a
resonant frequency of Wr = ᴨ / 2
if H (S) = S +0.1
______
(S +0.1) 2 +16

Sol: Ὡ = 4 & Wr = ᴨ/2 So T can be calculated T = 1/ 2 &

Put s = 4 ( 1-z -1/1+z -1 ) and

obtain H (Z) = 0.1289 + 0.006Z -1 + 0.122 Z -2

______________________
1 + 0.0006Z -1 + 0.97 Z -2
Example 4:

Design the analog filter into digital filter with a 3db

bandwidth of 0.25 ᴨ.

If H (S)= Ὡ/ S + Ὡ

Use Bilinear Transformation find H (Z)

Contd:
Advantages of bi-linear transformation method :
•The mapping is one to one
•There is no aliasing effect
•Stable analog filter is transformed into the stable digital filter
•There is no restriction one type of filter that can be transformed
•There is one to one transformation from the s-domain to the Z- domain
Disadvantages of bi-linear transformation method :
•The mapping is non linear in this method because of this frequency warping
effect takes place
 Frequency warping:

• This equation gives the relationship between the frequencies in the two domains
W = 2 tan -1 ῼ T (A)
—
2

It can be noted that the entire range in Ω is mapped only once into the range

• However, the mapping is non-linear and the lower frequencies in analog domain are expanded
in the digital domain, whereas the higher frequencies are compressed.
• The distortion introduced in the frequency scale of the digital filter to that of the analog filter
due to the non linearity of the arctangent function and this effect of the bilinear transform is
usually called frequency warping the filter design.
Prewarping:
• We can remove the warping problem using a simple technique.

• All we have to do to get a digital IIR filter with the same desired cutoff frequency as
ωC is to design an analog filter with a cutoff frequency that maps to ωC after
Bilinear Transformation.

We call this process Pre-warping.

We need to pre-warp the analog filters.

Prewarping:
 Digital Signal Processing
(BEC 503)
Unit-2
All-Pole Analog Filters
Dr. Neelesh Kumar Gupta
Professor
Department of Electronics and Communication Engineering
Ajay Kumar Garg Engineering College, Ghaziabad
IIR Filter Design

Designing techniques of IIR Filter

Approximation of Derivative Method

Impulse Invariant Method
Bi-Linear Transformation Method
(Frequency transformation is also possible)

Butterworth & Chebyshev Filter designing are there.

All Pole Analog Filters:
Butterworth Filter
The Butterworth low-pass filter has a magnitude response given by:

N=1,2,3…….

Fig.1 Low Pass Butterworth Filter response

Contd:

• Where, A is the filter gain and ωc is the 3 dB cut-off frequency and N is the
order of the filter. The magnitude response of the Butterworth filter is shown
in Fig. 1.

• The magnitude response has a maximally flat passband and stopband. It can
be seen that by increasing the filter order N, the Butterworth response
approximates the ideal response.

• However, the phase response of the Butterworth filter becomes more non-
linear with increasing N.
Properties of Butterworth filters
Steps of Designing

• To find analog pass & stop band frequency using the formula (As per used method)
• To find Order of the filter (N).
• To determine the analog cut off frequency (Ωc)
• To find the poles for Ha(s), Poles depends upon above mentioned parameters,
Pi= ± Ωc ej (N+2i+1) π/2N i= 0,1,2------N-1

Ha(s) = Ωc N
____________
(s-p ) (s-p )-----
1 2

Convert H(S) into H (Z) by using Impulse invariant or Bi-Linear transformation method.

In this way we can design filter Butterworth filter using these steps
If we use Impulse Invariant Method If we use Bi-Linear Transformation
Method

Cut Off Frequency

Order Of Filter
Example: Design a Butterworth II order LPF Using Bi-linear Transformation, if N=
2, Cut off freq. of analog filter= 1 KHz & sampling frequency is 10000 Hz.

Sol: At first continues frequency is to be converted into discrete frequency so

1000/10000 = 0.1 cycles/samples so

ωc = 2 ᴨ 0.1 = 0.2 ᴨ rad/sample & T = 1/10000

Ωc = 2/T tan ω /2 = 2 (10000) tan (0.2 ᴨ /2) = 6498.39 rad /sec

 Order of Filter N = 2
For poles using the formula by putting, (N=2)

Pi= ± Ωc e j(N+2i+1) π /2N i= 0,1,2------N-1

P 0 = ± 6498.39 (-0.707+j0.707)

P 1 = ± 6498.39 (-0.707-j0.707)
Ha(s) = Ωc N
____________
(s-p ) (s-p )-----
1 2
Finally, we will get

Ha (s) = (6498.39) 2
_________
s 2 +9190.1 s+ 42.22 10 6

Finally, H (Z) will come.

Butterworth Filter based examples
Example Design a Butterworth digital filter using the bilinear
transformation. The specifications of the desired low-pass filter are
given below with T = 1s
Contd:
Contd:
After Solving it we wil get Transfer Function of Analog Filter
Chebyshev Filters
The Chebyshev low-pass filter has a magnitude response given by
 Contd:
• The magnitude response of the Chebyshev filter is shown in Fig.
• The magnitude response has equiripple passband and maximally flat stopband.
• It can be seen that by increasing the filter order N, the Chebyshev response
approximates the ideal response. The phase response of the Chebyshev filter is more
non-linear than the Butterworth filter for a given filter length N.
Example: Design a digital Chebyshev filter to satisfy the constraints

using bilinear transformation and assuming T = 1 s.

Frequency Transformation

• There are basically four types of frequency selective filters, eg. low-pass, high-pass,
band-pass and band-stop.
• In the design techniques discussed so far we have considered only low-pass filters.
• This low-pass filter can be considered as a prototype filter and its system function can
be obtained.
• Then, if a high-pass or band-pass or band-stop filter is to be designed, it can be easily
obtained by using frequency transformation.
Frequency Transformation (cont.)
Frequency transformation can be accomplished in two ways.
(1) Analog frequency transformation
(2) Digital frequency transformation
 Frequency Transformation (cont.)
• In the analog frequency transformation, the analog system function Hp(s) of the
prototype filter is converted into another analog system function H(s) of the desired
filter.
• Then using any of the mapping techniques, it is converted into the digital filter having a
system function H(z).

Using frequency
Using mapping technique
transformation to convert
Prototype LPF HP(s) to convert into digital
into desired analog filter
filter
H(s)
 Frequency Transformation (cont.)

• In the digital frequency transformation, the analog prototype filter is first transformed
to the digital domain, to have a system function Hp(z).
• Then using frequency transformation, it can be converted into the desired digital filter.

Using frequency
Prototype LPF Prototype digital
transformation to
HP(s) filter HP(z)
get desired filter.
1. Analog Frequency Transformation
• The frequency transformation formulae used to convert a prototype low-pass filter into a
low-pass (with a different cut-off frequency), high-pass, band-pass or band-stop.
• Here Ωc is the cutoff frequency of the low-pass prototype filter.
• Ω*c cutoff frequency of new low-pass filter or high-pass filter and Ω1 and Ω2 are the
cutoff frequencies of band-pass or band-stop filters.
1. Analog Frequency Transformation
Summary
Example 1:
Example 1:
Example 2:
Example 2:
2. Digital Frequency Transformation
• As in the analog domain, frequency transformation is possible in the digital domain also.
• The frequency transformation is done in the digital domain by replacing the variable
𝑍 −1 by a function of 𝑍 −1 , i.e. f(𝑍 −1 ).
• This mapping must take into account the stability criterion.
• All the poles lying within the unit circle must map onto itself and the unit circle must also
map onto itself.
 Cont.
• For the unit circle to map onto itself, the implication is that for r = 1,

Hence, we must have |f(𝑒 −𝑗𝑤 )| = 1 for all frequencies. So, the mapping is that of an
all-pass filter and of the form
Summary
Summary
References

• John G Prokias, Dimitris G Manolakis, Digital Signal Processing. Pearson , 4th

Edition, 2007
• S. Salivahanan, A. Vallavaraj, Digital Signal Processing, McGraw Hill, 4th Edition
2017.
• Johnny R. Johnson, Digital Signal Processing, PHI Learning Pvt Ltd., 2009.
• S.K. Mitra, ‘Digital Signal Processing–A Computer Based Approach, McGraw
Hill, 4th Edition.
• Oppenheim & Schafer, Digital Signal Processing. Pearson Education 2015
THANK YOU