Scribd
Upload
98 views

Digital Signal Processing: Subject Code - ECN-312

This document discusses digital filter design, specifically IIR filter design. It covers various IIR filter design methods including impulse invariance, bilinear transformation, and designing analog prototypes. Impulse invariance directly maps analog filter coefficients to digital but has limitations. Bilinear transformation avoids aliasing by mapping the unit circle only once. Common analog filter types like Butterworth and Chebyshev are described. The document focuses on converting analog filter specifications to digital filter coefficients.

Uploaded by

Rajul Shakywar
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
98 views

Digital Signal Processing: Subject Code - ECN-312

This document discusses digital filter design, specifically IIR filter design. It covers various IIR filter design methods including impulse invariance, bilinear transformation, and designing analog prototypes. Impulse invariance directly maps analog filter coefficients to digital but has limitations. Bilinear transformation avoids aliasing by mapping the unit circle only once. Common analog filter types like Butterworth and Chebyshev are described. The document focuses on converting analog filter specifications to digital filter coefficients.

Uploaded by

Rajul Shakywar
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/ 40

Department of Electronics and Communication

Engineering

Digital Signal Processing

Subject Code – ECN-312

Digital filter Design (IIR)


Frequency Selective Response
Aim
Find appropriate coefficients of FIR , IIR filters
to approximate desired gain/phase characteristics
within specified frequency range.
Some General Observations
Requirement Preferred Filter
1. Linear Phase FIR
2. Lower Side-bands IIR
3.Less memory IIR
4. Fewer coefficients IIR Why
5. Lower complexity IIR
IIR Filter Design
• Better performance with less coefficients.
• Analog filter to digital conversion due to
mature field of Analog filters.

*Can be given by its


coefficients / its impulse
response
Stability condition while using these
conversions
Analog linear invariant system is stable , if its all
poles lie on left-half of the plane.

Methods should take care that the

Map

Map
IIR filters theoretically cannot have
linear phase
• Condition of linear phase in Z-domain

• All poles/zeros will have mirror image


counterpart along the radius of that pole/zero.
Stability condition will not satisfy!

The main focus is on magnitude characteristics! Phase is


related to the magnitude for a causal system.
Relation between digital and analog
frequency spectrums
OR
IIR Filter Design by impulse
invariance method
Sampling of analog filter impulse response:

Recalling the relation between spectrum of sampled


signal and spectrum of analog signal:
Impulse invariance method
• Aliasing occurs if the Fs is less than twice of
the highest frequency component of Xa.
Filter response of analog and digital filter
Observations
• If time period is very small then analog and
digital will be similar.
• Not appropriate for designing high-pass filters
due to aliasing components.

We need to investigate the relation of


(-j)

Now let us consider the mapping given by


Mapping for stable case

Mapping from analog to


digital domain is many to
one!

Digital domain frequency


lies between
Rearrange analog filter system function into
partial-fraction form

Use this h(n) for deciding filter


expression in digital domain:
This method will be successful for distinctive poles !
Example:
OR
Digital

Repeat this
problem in MATLAB
with different T
values.

Analog
Bilinear Transformation
• Impulse invariance method is limited to low-
pass and small cases of band-pass filters.
• Bilinear transformation maps the unit circle
only once, avoiding aliasing of components.
• Derivation starts by considering trapezoidal
rules for integration, lets start by considering
Trapezoidal rule in discrete domain
Design sequence in bi-linear
transformation
• Selecting frequency in digital domain.
• Specification are converted to analog domain

• Analog filter is designed.


• Digital filter is achieved:
Can be chosen
Designing analog filter is of prime interest
Common Analog Filters
Poles will be given as:
For a given frequency and attenuation at that frequency can
be easily given by:
Calculation
of order N!
Required parameter:
Chebyshev Filters
• Two types:
(a) Type-1 Chebyshev
(b) Type-2 Chebyshev

Chebyshev type-1:
Recursive Relation:
where
Method for locating poles

For Type-II Chebyshev:


Zeros:

Poles:
New poles are related to the poles of type-1 Chebyshev and
ripple parameter is defined as:

Required:

You might also like