Scribd
Upload
182 views5 pages

Digital Filter Design

This document discusses methods for calculating digital filter coefficients. It begins by designing a Hamming window FIR filter to given specifications and calculating its ideal and actual impulse responses. It then examines the magnitude response of the filter. Next, it demonstrates designing an FIR lowpass filter using the FIR1 function for different filter lengths. Finally, it discusses designing IIR filters using Butterworth, Chebyshev type 1 and 2, and Elliptic filters and examining their frequency and pole-zero responses.

Uploaded by

abhilash_bec
Copyright
© Attribution Non-Commercial (BY-NC)
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
182 views5 pages

Digital Filter Design

This document discusses methods for calculating digital filter coefficients. It begins by designing a Hamming window FIR filter to given specifications and calculating its ideal and actual impulse responses. It then examines the magnitude response of the filter. Next, it demonstrates designing an FIR lowpass filter using the FIR1 function for different filter lengths. Finally, it discusses designing IIR filters using Butterworth, Chebyshev type 1 and 2, and Elliptic filters and examining their frequency and pole-zero responses.

Uploaded by

abhilash_bec
Copyright
© Attribution Non-Commercial (BY-NC)
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/ 5

Untitled

1/5

Back
This page show simple Digital filter coefficient calculation method
[Hamming ] [FIR Filter] [IIR Filter]
[Hamming ]

1. Filter Design Using Hamming window satisfying below Specification.


wp
ws
tr_width
M

= 0.2*pi;
% Passband Frequency
= 0.3*pi;
% Stopband Frequency
= ws - wp;
% Transition Band Frequency
= ceil(6.6*pi/tr_width) + 1 % M = 67(filter order) , w = 6.6*pi / tr_width
% CEIL(X) rounds the elements of X to the nearest integers

towards infinity.
%
%
%
%
n
wc
hd
M-1

Rectangular(tr_width = 1.8pi/M) , Bartlett(tr_width = 6.1pi/M)


Blackman(tr_width = 11pi/M) , Hanning(tr_width = 6.2pi/M)
Hamming(tr_width = 6.6pi/M)
tr_width => M.

= [ 0 : 1 : M-1 ];
= (ws + wp) / 2
= ideal_lp(wc,M);

% Ideal LPF cutoff frequency


% hd = ideal impulse response between 0 to

subplot(2,2,1); stem(n,hd); title('Ideal Impulse Response');


axis([0 M-1 -0.1 0.3]); xlabel('n'); ylabel('hd(n)')
/****************** Ideal+-Low Pass Filter ****************************/
function hd = ideal_lp(wc,M);
% Ideal LowPass filter computation, [hd] = ideal_lp(wc,M)% hd = ideal impulse
response between 0 to M-1,
% wc = cutoff frequency in radians, M = length of the ideal filter
alpha = (M-1)/2;
n
= [0:1:(M-1)];
m
= n - alpha + eps;
% eps = 2.2204e-016
hd
= sin(wc*m) ./ (pi*m);
/***************************************************************************/
w_ham = (hamming(M))';
% w_ham = windows function
subplot(2,2,2), stem(n,w_ham); title('Hamming Window')
axis([0 M-1 0 1.1]); xlabel('n'); ylabel('w(n)')

C:\pec_works\PEC_Office\PEC_2008\Homepage\Naver_Homepage\doc\matlab.htm
2008-11-13 / 4:44

Untitled

2/5

h = hd .* w_ham; subplot(2,2,3); stem(n,h);


title('Actual Impulse Response')
axis([0 M-1 -0.1 0.3]); xlabel('n'); ylabel('h(n)')
[db,mag,pha,grd,w] = freqz_m(h,[1]);
%[db,mag,pha,grd,w] = freqz_m(hd,[1]);
%[db,mag,pha,grd,w] = freqz_m(w_ham,[1]);
delta_w = 2*pi/1000;
Rp
= -(min(db(1:1:wp/delta_w+1)))
% Actual Passband
Ripple
As
= -round(max(db(ws/delta_w+1:1:501)))
% Min Stopband
attenuation
subplot(2,2,4); plot(w/pi,db);title('Magnitude Response in dB');grid
axis([0 1 -100 10]); xlabel('frequency in pi units'); ylabel('Decibels')
/****************** freqz_m function define
***************************************/
function [db,mag,pha,grd,w] = freqz_m(b,a);
if (a =1) Fir Filter
% [db,mag,pha,grd,w] = freqz_m(b,a);
% db = Relative magnitude in dB computed over 0 to pi radians
% mag = absolute magnitude computed over 0 to pi radians
% pha = Phase response in radians over 0 to pi radians
% grd = Group delay over 0 to pi radians
%w
= 501 frequency samples between 0 to pi radians
%b
= numerator polynomial of H(z) (for FIR: b=h)
%a
= denominator polynomial of H(z) (for FIR: a=[1])
[H,w]
H
mag
db
pha
% pha
grd
% grd

= freqz(b,a,1000,'whole');
= (H(1:1:501))'; w = (w(1:1:501))';
= abs(H);
= 20*log10((mag+eps)/max(mag));
= angle(H);
= unwrap(angle(H));
= grpdelay(b,a,w);
= di

ff(pha);
% grd = [grd(1) grd];
% grd = [0 grd(1:1:500); grd; grd(2:1:501) 0];
% grd = median(grd)*500/pi;
/*************************************************************************************/
% filter length M = 67 , 52dB,

Pass Band Ripple = 0.0394

[Hamming] [IIR Filter] , [FIR Filter]

EX2 )

Fir method - LPF

C:\pec_works\PEC_Office\PEC_2008\Homepage\Naver_Homepage\doc\matlab.htm
2008-11-13 / 4:44

Untitled
clf
t
= 0 : 0.001 : 1;
= 1000
y
= sin(2*pi*30*t)+sin(2*pi*80*t)+cos(2*pi*160*t);
subplot(2,1,1), plot(y,'g'); title('time domain');

3/5

% N = (f) = 1/0.001

YY
= fft(y);
%
plot(abs(YY))
ylen
= length(YY);
fs
= 1000;
f1
= 1 : 1 : ylen/2 ;
ff
= ((f1*fs)/(ylen));
magYY
= abs(YY);
figure(1);
ylen2
= round(ylen/2-0.5);
subplot(2,1,2), plot(ff(1:ylen2), magYY(1:ylen2),'k'); title('frequency
domain');
hold on;
fn
= fs/2;
% fs =1000 ;
f1n
= 55/fn;
% fn = 500 , wc = 0.11, lowpass filter
cutoff frequency (30 + 80)/2
for k
= 1 : 5 : 100
[B]
= fir1(k,f1n);
% LPF , k = 0 ~ 100 ,f1n = 0.1100 ;
B:
filter coefficient
[H,wT]
= freqz(B,1,100); % B / 1 , N point
T
= 0.001;
herzt
= wT/(2*pi*T);
figure(1);
subplot(2,1,1), plot(y,'b'); title('time domain , input data display ');
subplot(2,1,2), plot(ff(1:ylen2),magYY(1:ylen2),'k'); title('frequency domain,
fft(frequency analysis');
hold on;
s
= int2str(k);
subplot(2,1,2), plot(herzt,abs(H)* 600,'r');title('in fft plot,Lowpass
fillter'),xlabel('Frequence(Hz)');
ylabel('Amplitude'); text( 250 + k * 8,500,s ); grid; axis([0 1000 -1 650]);
%hold off;
hold on;
pause;
figure(2)
clf;
result = filter(B,1,y);
% y = input data , B = filter coefficient ,
% result = frequency domain filter shape.
plot(result,'c');
plot( result(300:500) );
result
end

C:\pec_works\PEC_Office\PEC_2008\Homepage\Naver_Homepage\doc\matlab.htm
2008-11-13 / 4:44

Untitled
[Hamming] [FIR Filter] ,

4/5
[IIR Filter]

3. IIR Digital Filter


Define of formal - LPF ,HPF Example of butter / cheby1,2 / ellips
t
= 0:0.001:1-0.001;
y
= sin(2*pi*30*t)+sin(2*pi*80*t)+cos(2*pi*160*t);
YY
= fft(y);
ylen = length(YY);
fs
= 1000;
f1
= 1:1:ylen;
ff
= ((f1*fs)/(ylen));
magYY = abs(YY);
hold on;
for k = 1 : 2 : 17
% input frequency ; 30Hz, 80Hz, 160Hz
% [B,A] = butter(k,0.12);
% [B,A]= BUTTER(N,W) LPF ,Wn = 0.12 (W = 0 ~
0.5) ,BFP ( w1 < W < w2 )
% [B,A] = BUTTER(k,0.24,'high') % [B,A] = BUTTER(N,Wn,'high') - highpass
filter, 0.24 = 120Hz / (1000Hz/2)
% butter filter = 11 order saturation start , tr_width is width(disadvantage)
% [B,A] = CHEBY1(k,0.5,0.12)
% [B,A] = CHEBY1(N,R,Wn) N order, passband
ripple R(0.5),Wn(cut off frequency,0 ~ 1)
% [B,A] = cheby1(k,0.5,0.24,'high'); % Chebyshev type I Rp = 0.5 dB(starting
point)
% cheby1 filter = 7 order saturation start , tr_width is not width
% [B,A] = cheby2(k,20,0.12);
% Chebyshev type 2 Rp = 20 dB(starting point)
% [B,A] = cheby2(k,20,0.24,'high');
% Chebyshev type 2 Rp = 20 dB(starting
point)
% cheby2 5 order saturation start, stop band ripple is high
% [B,A] = ELLIP(k, 3, 30 ,0.12) % [B,A] = ELLIP(N,Rp,Rs,Wn),[B,A] =
ELLIP(N,Rp,Rs,Wn,'high') or 'stop' or 'W'
[B,A] = ELLIP(k,3, 30,0.24,'high')
% Ellip 3 order saturation start , Just ripple control is difficult in manual haddle.
[H,wT]= freqz(B,A,100);
T
= 0.001;
herzt = wT/(2*pi*T);
figure(1);
subplot(2,1,1),plot(y,'g'),title('time domain');
subplot(2,1,2),plot(ff(1:ylen/2),magYY(1:ylen/2),'k'),title('frequenc
y domain');
hold on;
subplot(2,1,2),plot(herzt,abs(H)*700,'r'); title('Low fillter'),grid; pause

C:\pec_works\PEC_Office\PEC_2008\Homepage\Naver_Homepage\doc\matlab.htm
2008-11-13 / 4:44

Untitled

5/5

s = int2str(k) ;text(250 + k * 8,700,s) ; grid ; axis([0 1000 -1 650]) ;


figure(2) ;
subplot(2,1,1); result = filter(B,A,y); plot(result,'c');
subplot(2,1,2) ;plot(result(300:500));
figure(3) ;
zplane(B,A) ;title('Pole-Zero Plot');
%A = [ 1 1/3], B = [ 1 -1/2] , zplane(Z(B),P(A))
% H(e) = B(zeros) / A(poles)
% Filter(B(Numberator- ZEROS),A(denominator - POLES))
% [H,W] = FREQZ(B(Numberator -zeros),A(denominator - Poles),N)
end

C:\pec_works\PEC_Office\PEC_2008\Homepage\Naver_Homepage\doc\matlab.htm
2008-11-13 / 4:44

You might also like