Design of Adaptive Filter Based on Matlab and Simulink - Advanced Research on Computer Education, Simulation and Modeling - page 295

noise n from input signal by signal ( )

xn

= sin(100 π

ft

) , we can filter noise signal ( )

adaptive notch filter.
In Fig.2, we make output signal

approximate in least mean square error y ( n )

d

( n )

by adjusting adaptively filter weights

w and w . At this time, error signal

is e

( n )

output signal of adaptive notch filter.
(1) The simulation of adaptive notch filter in Matlab
%main
…
s=10*sin(4*pi*t)+3*sin(100*pi*t);
…
x1= sin(100*pi*t);
x2= sin(100*pi*t+pi/2);
…
for k=1:50
y(k)= w1(k)*x1(k)+ w2(k)*x2(k);
e(k)=s(k)-y(k);
w1(k+1)=w1(k)+2*0.1*e(k)*x1(k);
w2(k+1)=w2(k)+2*0.1*e(k)*x2(k);
end
…
(2) The simulation of adaptive notch filter in Simulink
We can establish the corresponding Simulink simulation model according to the
principle diagram of adaptive notch filter of Fig.2, as shown in Fig. 3.

Fig. 3. The simulation model of adaptive notch filter based on LMS algorithm