Abhinav Arora (09)

Practical No: -07

Aim: To write a program to design Butterworth filter using bilinear transformation

Theory:

IIR FILTERS
An infinite impulse response (IIR) filter is a digital filter that depends linearly on a finite number of input
samples and a finite number of previous filter outputs. In other words, it combines a FIR filter with
feedback from previous filter outputs. Mathematically, for some coefficients bi and aj:
The infinite impulse response (IIR) filter is a recursive filter in that the output from the filter is computed
by using the current and previous inputs and previous outputs.
Because the filter uses previous values of the output, there is feedback of the output in the filter structure.
The design of the IIR filter is based on identifying the pulse transfer function G(z) that satisfies the
requirements of the filter specification. This can be undertaken either by developing an analogue
prototype and then transforming it to the pulse transfer function, or by designing directly in digital. Figure
shows typical IIR filter architecture.

Infinite Impulse Response Filter Design

The most commonly used IIR filter design method uses reference analog prototype filter. It is the best
method to use when designing standard filters such as low-pass, high-pass, bandpass and band-stop
filters.

The filter design process starts with specifications and requirements of the desirable IIR filter. A type of
reference analog prototype filter to be used is specified according to the specifications and after that
everything is ready for analog prototype filter design.
The next step in the design process is scaling of the frequency range of analog prototype filter into
desirable frequency range. This is how an analog prototype filter is converted into an analog filter.
After the analog filter is designed, it is time to go through the last step in the digital IIR filter design
process. It is conversion from analog to digital filter. The most popular and most commonly used
converting method is bilinear transformation method. The resulting filter, obtained in this way, is always
stable. However, instability of the resulting filter, when bilinear transformation is used, may be caused
only by the finite word-length side-effect.
Abhinav Arora (09)

IIR Filters Types:

1) Butterworth IIR Filter

2) Chebyshev IIR Filter

3) Bessel IIR Filter

4) Elliptic IIR Filter

Transformation Methods – Analog IIR Filter Transfer Function [ H(S)] to Digital IIR Filter Transfer
Function [ H(Z)]

1) Bilinear Transformation

2) Impulse Invariant Method

3) Matched Z Transform

4) Approximation of Derivatives Method

Abhinav Arora (09)

Butterworth IIR Filter

Butterworth, Caur, and Chebyshev are some of the most commonly used filters that can provide a near-
ideal response curve. In them, we will discuss the Butterworth filter here as it is the most popular one of
the three.

The main features of the Butterworth filter are:

 It is an R-C (Resistor, Capacitor) & Op-amp (operational amplifier) based filter

 It is an active filter so the gain can be adjusted if needed
 The key characteristic of Butterworth is that it has a flat passband and flat stopband. This is the
reason it is usually called ‘flat-flat filter’.
Now let us discuss the circuit model of Low Pass Butterworth Filter for a better understanding.

First Order Low Pass Butterworth Filter

The figure shows the circuit model of the first-order low-pass Butter worth filter.

In the circuit we have:

 Voltage ‘Vin’ as an input voltage signal which is analog in nature.

 Voltage ‘Vo’ is the output voltage of the operational amplifier.
 Resistors ‘RF’ and ‘R1’ are the negative feedback resistors of the operational amplifier.
 There is a single R-C network (marked in the red square) present in the circuit hence the filter is a
first-order low pass filter
 ‘RL’ is the load resistance connected at the op-amp output.
Abhinav Arora (09)

If we use the voltage divider rule at point ‘V1’then we can get the voltage across the capacitor as,

V1 = [ -jXc / (R-jXc) ] Vin Here –jXc = 1/2ᴫfc

After substitution this equation we will have something like below

V1 = Vin / (1+j2ᴫfRC)

Now the op-amp here used in negative feedback configuration and for such a case the output voltage
equation is given as,

V0 = ( 1 + RF / R1 ) V1

This is a standard formula and you can look into op-amp circuits for more details.

If we submit V1 equation into Vo we will have,

V0 = (1 + RF / R1) [Vin / (1 + j2ᴫfRC) ]

After rewriting this equation we can have,

V0 / Vin = AF / ( 1 + j(f/fL) )

In this equation,

 V0 / Vin = gain of the filter as a function of frequency

 AF = (1+RF / R1) = passband gain of the filter
 f = frequency of the input signal
 fL = 1 / 2ᴫRC = cutoff frequency of the filter. We can use this equation to choose appropriate
resistor and capacitor values to select the cutoff frequency of the circuit.
Abhinav Arora (09)

Second-Order Butterworth Low Pass Filter

The figure shows the circuit model of the 2nd order Butterworth low pass filter.

In the circuit we have:

 Voltage ‘Vin’ as an input voltage signal which is analog in nature.

 Voltage ‘Vo’ is the output voltage of the operational amplifier.
 Resistors ‘RF’ and ‘R1’ are the negative feedback resistors of the operational amplifier.
 There is a double R-C network (marked in a red square) present in the circuit hence the filter is a
second-order low pass filter.
 ‘RL’ is the load resistance connected at the op-amp output.

Second Order Low Pass Butterworth Filter Derivation 

Second-order filters are important because higher-order filters are designed using them. The gain of the
second-order filter is set by R1 and RF, while the cutoff frequency fH is determined by  R2, R3, C2 &
C3 values. The derivation for the cutoff frequency is given as follows,

fH = 1 / 2ᴫ(R2R3C2C3)1/2

The voltage gain equation for this circuit can also be found in a similar way as before and this equation is
given below,
Abhinav Arora (09)

In this equation,

 V0 / Vin = gain of the filter as a function of frequency

 AF = (1 + RF/R1) passband gain of the filter
 f = frequency of the input signal
 fH = 1 / 2ᴫ(R2R3C2C3)1/2 = cutoff frequency of the filter. We can use this equation to choose
appropriate resistor and capacitor values to select the cutoff frequency of the circuit. Also if we choose
the same resistor and capacitor in the R-C network then the equation becomes,

We can the voltage gain equation to observe the change in gain magnitude with the corresponding change
in the frequency of the input signal.

Bilinear Transformation: -

The bilinear transformation is a mathematical mapping of variables. In digital filtering, it is a standard

method of mapping the s or analog plane into the z or digital plane. It transforms analog filters, designed
using classical filter design techniques, into their discrete equivalents.

The bilinear transform is defined by the substitution

(8.6)
   (typically)

(8.7)

where   is some positive constant That is, given a continuous-time transfer function  , we apply the
bilinear transform by defining
Abhinav Arora (09)

where the `` '' subscript denotes ``digital,'' and `` '' denotes ``analog.''

It can be seen that analog dc ( ) maps to digital dc ( ) and the highest analog frequency (
) maps to the highest digital frequency ( ). It is easy to show that the entire   axis in

the   plane (where  ) is mapped exactly once around the unit circle in the   plane (rather than
summing around it infinitely many times, or ``aliasing'' as it does in ordinary sampling). With   real and
positive, the left-half   plane maps to the interior of the unit circle, and the right-half   plane maps
outside the unit circle. This means stability is preserved when mapping a continuous-time transfer
function to discrete time.

Another valuable property of the bilinear transform is that order is preserved. That is, an  th-order  -
plane transfer function carries over to an  th-order  -plane transfer function. (Order in both cases
equals the maximum of the degrees of the numerator and denominator polynomials

The constant   provides one remaining degree of freedom which can be used to map any particular finite
frequency from the   axis in the   plane to a particular desired location on the unit circle   in
the   plane. All other frequencies will be warped. In particular, approaching half the sampling rate, the
frequency axis compresses more and more. Note that at most one resonant frequency can be preserved
under the bilinear transformation of a mass-spring-dashpot system. On the other hand, filters having a
single transition frequency, such as lowpass or high pass filters, map beautifully under the bilinear
transform; one simply uses   to map the cut-off frequency where it belongs, and the response looks great.
In particular, ``equal ripple'' is preserved for optimal filters of the elliptic and Chebyshev types because
the values taken on by the frequency response are identical in both cases; only the frequency axis is
warped. The bilinear transform is often used to design digital filters from analog prototype filters.
Abhinav Arora (09)

Bilinear Transformation simple Example--:

Design of Digital Butterworth IIR Filter using Bilinear Transformation.

Procedure:

Given Parameters in problem

wp – Digital Pass band Edge Frequency δ1 = Pass band Ripple

ws –Digital Stop band Edge Frequency δ2 = Pass band Ripple

Step 1:

Ωp = Analog Pass band Edge Frequency = (2/T) tan( wp / 2)

Ωs = Analog Pass band Edge Frequency = (2/T) tan( ws / 2)

Step 2:

Filter Order =

1
( )
( )
−1
δ 2
2
log
1

N≥
( δ 2
1
−1
)
Ωs
2 log ( ΩP )
Abhinav Arora (09)

Step 3:

Analog Cut off Frequency Ωc = Ωp / [(1/ δ12) -1] (1/2N)

Step 4:

Analog Butterworth IIR Filter transfer function

(Ωc) N
H(S) = ---------
poles

Step 5:

Poles

S= j Ωc e j (2 k+1)π /2 N

Where k =0,1,2…. N-1

Step 6:

Digital IIR Butterworth Filter from Analog IIR Butterworth Filter

(1-z-1)
H(Z) = H(S) / S = (2/T) -------
(1+z-1)
Abhinav Arora (09)

Bilinear Transform Lowpass Butterworth Filter Design Ex.

Bilinear Transform Lowpass Butterworth Filter Design Ex-:

Abhinav Arora (09)
Abhinav Arora (09)

Program to design Butterworth filter using bilinear transformation in

MATLAB:

clc ;

close all;

clear all; 

b=input('Enter numerator coefficient of analog filter:')

a=input('Enter denominator coefficient of analog filter:')

f=input('Enter sampling frequency:')

Ts=1/f

[bz,az]=bilinear(b,a,f)

disp('Equivalent digital transfer function H(Z)is =')

systf=tf(bz,az,Ts)

 figure(1)

freqz(bz,az)

figure(2)

zplane(bz,az)

 disp('The residues, pole locations and the direct terms are:') ;

[R,p,c]=residuez(bz,az)
Abhinav Arora (09)

OUTPUT:

Enter numerator coefficient of analog filter:2

b =     2

Enter denominator coefficient of analog filter:[1 3 2]

a =     1     3     2

Enter sampling frequency:1

f =     1
Ts =    1
bz =    0.1667    0.3333    0.1667
az =    1.0000   -0.3333    0.0000

Equavilent digital transfer function H(Z)is =

systf =  0.1667 z^2 + 0.3333 z + 0.1667

  ------------------------------
     z^2 - 0.3333 z + 7.401e-17

Sample time: 1 seconds

Discrete-time transfer function.

The residues, pole locations and the direct terms are:

R =   1.0e+15 *
    0.0000
    -2.2518

p =    0.3333
    0.0000

c =    2.2518e+15
Abhinav Arora (09)