# How can I apply median filter with sliding window for the ECG signal?

19 views (last 30 days)
Dhiyaa Al-Shammari on 25 Nov 2019
Commented: Daniel M on 25 Nov 2019
How can I apply median filter with sliding window n=200 ms and n=600 ms in Matlab fo a signalat sample rate 360 Hz?
figure(1);
plot(val)
x=medfilt1(val,200);
figure(2);
plot(x)
y=medfilt1(val,600);
figure(3);
plot(y)

Daniel M on 25 Nov 2019
Edited: Daniel M on 25 Nov 2019
The second input of the function medfilt1 determines the window size.
% Y = MEDFILT1(X,N) specifies the order, N, of the median filter.
% For N odd, Y(k) is the median of X( k-(N-1)/2 : k+(N-1)/2 ).
% For N even, Y(k) is the median of X( k-N/2 : k+N/2-1 ).
In your case, the order will be however many samples it takes to 200 ms at 360 Hz. E.g. 0.600*fs = 216. Otherwise, it will be length(t)-1.
fs = 360; % Hz
t200 = 0:1/fs:0.2;
t600 = 0:1/fs:0.6;
figure(1);
plot(val)
x=medfilt1(val,length(t200)-1);
figure(2);
plot(x)
y=medfilt1(val,length(t600)-1);
figure(3);
plot(y)
##### 2 CommentsShowHide 1 older comment
Daniel M on 25 Nov 2019
x is already a median filtered version of val. If you want to filter it again, then use x as an input to medfilt1.