Butterworth Filter Design

Table of Contents
Title: Butterworth Filter Design................................................................................................................................. 1
Student Info:............................................................................................................................................................. 1
Experiment no. : 8.............................................................................................................................................1
AIM: ......................................................................................................................................................................... 1
SOFTWARE USED:..............................................................................................................................................2
THEORY:.................................................................................................................................................................. 2
RESULTS:................................................................................................................................................................ 3
Example 1.................................................................................................................................................................3
Step 1: ................................................................................................................................................................. 3
Step 2:.................................................................................................................................................................. 3
Step 3:.................................................................................................................................................................. 4
Step 4:.................................................................................................................................................................. 5
Example 2.................................................................................................................................................................6
Step 1: ................................................................................................................................................................. 6
Step 2:.................................................................................................................................................................. 6
Step 3:.................................................................................................................................................................. 6
Step 4:.................................................................................................................................................................. 7
CONCLUSION:.........................................................................................................................................................9

Title: Butterworth Filter Design

Student Info:
Name: Ayush Meshram

Uid:2022200081

Batch:EXTC B-1

Experiment no. : 8

AIM:

1
SOFTWARE USED:
MATLAB ONLINE

THEORY:
Butterworth Filters

The amplitude response (Hjw)| of an nth order Butterworth lowpass filter is given by

Observe that at W = 0, the gain [H (jo)| is unity and at W = Wc, the gain |H (jw.)| = 1/1.414 or -3 dB. The gain
drops by a factor 1.414 at W = Wc . Because the power is proportional to the amplitude squared, the power
ratio (output power to input power) drops by a factor 2 at W = Wc. For this reason Wc is called the half-power
frequeney or the 3 dB-cutoff frequency (amplitude ratio of V2 is 3 dB).

The amplitude response |H(jw)| of the normalized lowpass Butterworth filters is depicted in Fig. 7.20 for various
values of n. From Fig. 7.20

2
we observe the following:

1. The Butterworth amplitude response decreases monotonically. Moreover, the first 2n - 1 derivatives of the
amplitude response are zero at w = 0. For this reason this characteristic is called maximally fat at w = 0.
Observe that a constant characteristic (ideal) is maximally flat for all w < 1. In the Butterworth filter we try to
retain this property at least at the origin.†

2. The filter gain is 1 (0 dB) at w = 0 and 0.707 (-3 dB) at w = 1 for all n. Therefore, the 3-dB (or half power)
bandwidth is 1 rad/s for all n.

3. For large n, the amplitude response approaches the ideal characteristic. To determine the corresponding
transfer function H(s), recall

Normalized Filter

In the design procedure it proves most convenient to consider a normalized filter H(s), whose half-power
frequency is 1 rad/s (wc = 1). For such a filter, the amplitude characteristic reduces to

RESULTS:

Example 1
Step 1:
Determine n

Wp = 10; % Passband frequency

Ws = 20; % Stopband frequency
Gp = -2; % Passband gain
Gs = -20; % Stopband gain

% Compute passband and stopband attenuation in linear scale

Ps=-Gs/10; Pp= -Gp/10; Wsp=Ws/Wp; %Defining variable

% Compute filter order

nc = (log((10.^(Ps) -1)/( 10.^(Pp) - 1))) / (2.*log(Wsp))

nc = 3.7016

n = ceil(nc)

n = 4

Step 2:
Determine Wc

3
 % Compute cutoff frequency
Wc = (Wp) / (10.^(Pp) - 1).^(1/(2.*n))

Wc = 10.6934

% Compute stopband cutoff frequency

Wc1 = (Ws) / (10.^(Ps) -1).^(1/(2.*n))

Wc1 = 11.2610

Step 3:
Determine the normalized transfer function

s = []

s =

[]

for k=1:n
A = (2.*(k-1)+n+1) / (2.*n)
Sk = cos(A.*pi) + 1i*sin(A.*pi)
s = [s Sk]
end

A = 0.6250
Sk = -0.3827 + 0.9239i
s = -0.3827 + 0.9239i
A = 0.8750
Sk = -0.9239 + 0.3827i
s = 1×2 complex
-0.3827 + 0.9239i -0.9239 + 0.3827i
A = 1.1250
Sk = -0.9239 - 0.3827i
s = 1×3 complex
-0.3827 + 0.9239i -0.9239 + 0.3827i -0.9239 - 0.3827i
A = 1.3750
Sk = -0.3827 - 0.9239i
s = 1×4 complex
-0.3827 + 0.9239i -0.9239 + 0.3827i -0.9239 - 0.3827i -0.3827 - 0.9239i

s = s'

s = 4×1 complex
-0.3827 - 0.9239i
-0.9239 - 0.3827i
-0.9239 + 0.3827i
-0.3827 + 0.9239i

num1 = [0 1]

num1 = 1×2
0 1

den1 = poly(s)

den1 = 1×5 complex

1.0000 + 0.0000i 2.6131 + 0.0000i 3.4142 + 0.0000i 2.6131 + 0.0000i

4
Step 4:
Determine the final filter transfer function

% Display filter order and cutoff frequency

disp(['Filter Order is n = ' num2str(n)]);

Filter Order is n = 4

disp(['Cutoff Frequency of the Filter is Wc ' num2str(Wc)]);

Cutoff Frequency of the Filter is Wc 10.6934

% Display poles of the transfer function

disp('Poles of the transfer function are');

Poles of the transfer function are

for k = 1:n
disp([num2str(real(s(k))) ' ' num2str(imag(s(k)))]);
end

-0.38268 -0.92388
-0.92388 -0.38268
-0.92388 0.38268
-0.38268 0.92388

den2= poly(Wc.*s)

den2 = 1×5 complex

104 ×
0.0001 + 0.0000i 0.0028 + 0.0000i 0.0390 + 0.0000i 0.3195 + 0.0000i

num2 = [0 Wc^n]

num2 = 1×2
104 ×
0 1.3076

disp("The Normalised fourth order transfer function is: ")

The Normalised fourth order transfer function is:

printsys(abs(num1),abs(den1))

num/den =

1
--------------------------------------------
s^4 + 2.6131 s^3 + 3.4142 s^2 + 2.6131 s + 1

disp("The transfer function with s replaced by s/Wc: ")

The transfer function with s replaced by s/Wc:

5
 printsys(abs(num2),abs(den2))

num/den =

13075.6027
-----------------------------------------------------------
s^4 + 27.9432 s^3 + 390.4105 s^2 + 3195.2631 s + 13075.6027

Example 2
Determine n, the order of the lowpass Butterworth filter to meet the following specificationsWp =
100 ;Ws = 200 ; Gp = -0.5dB ; Gs = -20dB

Step 1:
Determine n

Wp = 100; % Passband frequency

Ws = 200; % Stopband frequency
Gp = -0.5; % Passband gain
Gs = -20; % Stopband gain

% Compute passband and stopband attenuation in linear scale

Ps=-Gs/10; Pp= -Gp/10; Wsp=Ws/Wp; %Defining variable

% Compute filter order

nc = (log((10.^(Ps) -1)/( 10.^(Pp) - 1))) / (2.*log(Wsp))

nc = 4.8321

n = ceil(nc)

n = 5

Step 2:
Determine Wc

% Compute cutoff frequency

Wc = (Wp) / (10.^(Pp) - 1).^(1/(2.*n))

Wc = 123.4120

% Compute stopband cutoff frequency

Wc1 = (Ws) / (10.^(Ps) -1).^(1/(2.*n))

Wc1 = 126.3184

Step 3:
Determine the normalized transfer function

s = []

6
 s =

[]

for k=1:n
A = (2.*(k-1)+n+1) / (2.*n)
Sk = cos(A.*pi) + 1i*sin(A.*pi)
s = [s Sk]
end

A = 0.6000
Sk = -0.3090 + 0.9511i
s = -0.3090 + 0.9511i
A = 0.8000
Sk = -0.8090 + 0.5878i
s = 1×2 complex
-0.3090 + 0.9511i -0.8090 + 0.5878i
A = 1
Sk = -1.0000 + 0.0000i
s = 1×3 complex
-0.3090 + 0.9511i -0.8090 + 0.5878i -1.0000 + 0.0000i
A = 1.2000
Sk = -0.8090 - 0.5878i
s = 1×4 complex
-0.3090 + 0.9511i -0.8090 + 0.5878i -1.0000 + 0.0000i -0.8090 - 0.5878i
A = 1.4000
Sk = -0.3090 - 0.9511i
s = 1×5 complex
-0.3090 + 0.9511i -0.8090 + 0.5878i -1.0000 + 0.0000i -0.8090 - 0.5878i

s = s'

s = 5×1 complex
-0.3090 - 0.9511i
-0.8090 - 0.5878i
-1.0000 - 0.0000i
-0.8090 + 0.5878i
-0.3090 + 0.9511i

num1 = [0 1]

num1 = 1×2
0 1

den1 = poly(s)

den1 = 1×6 complex

1.0000 + 0.0000i 3.2361 + 0.0000i 5.2361 + 0.0000i 5.2361 + 0.0000i

Step 4:
Determine the final filter transfer function

% Display filter order and cutoff frequency

disp(['Filter Order is n = ' num2str(n)]);

Filter Order is n = 5

disp(['Cutoff Frequency of the Filter is Wc ' num2str(Wc)]);

7
Cutoff Frequency of the Filter is Wc 123.412

% Display poles of the transfer function

disp('Poles of the transfer function are');

Poles of the transfer function are

for k = 1:n
disp([num2str(real(s(k))) ' ' num2str(imag(s(k)))]);
end

-0.30902 -0.95106
-0.80902 -0.58779
-1 -1.2246e-16
-0.80902 0.58779
-0.30902 0.95106

den2= poly(Wc.*s)

den2 = 1×6 complex

1010 ×
0.0000 + 0.0000i 0.0000 + 0.0000i 0.0000 + 0.0000i 0.0010 + 0.0000i

num2 = [0 Wc^n]

num2 = 1×2
1010 ×
0 2.8628

disp("The Normalised fourth order transfer function is: ")

The Normalised fourth order transfer function is:

printsys(abs(num1),abs(den1))

num/den =

1
---------------------------------------------------------
s^5 + 3.2361 s^4 + 5.2361 s^3 + 5.2361 s^2 + 3.2361 s + 1

disp("The transfer function with s replaced by s/Wc: ")

The transfer function with s replaced by s/Wc:

printsys(abs(num2),abs(den2))

num/den =

28627751612.4319
-----------------------------------------------------------------------
s^5 + 399.3697 s^4 + 79748.0683 s^3 + 9841869.9143 s^2 + 750667179.66 s
+ 28627751612.4319

8
CONCLUSION: