IIR filter as amoving averege

2 Ansichten (letzte 30 Tage)
Rica
Rica am 22 Mär. 2013
hi! I want to use the filter function as a moving averege filter. the Type of the filter should be IIR Filter. Te window length is 3. for exemple:
%
X=rand(1:300)
B=1/3
A=1-B;
Y=filter(B,A,X)->First order IIR filter
the filter functi should start after reaching the window lenght. for the first 2 value (before reaching the window lenght of 3) , i wan to calculate just the mean value.
%
Y(1)=X(1)
Y(2)=mean(Y(1)+X(2));
Y(3:end)=filter(B,A,X(3:end))
could someone help me ?
thank you

Akzeptierte Antwort

Andrei Bobrov
Andrei Bobrov am 22 Mär. 2013
Bearbeitet: Andrei Bobrov am 22 Mär. 2013
EDIT
out = filter([1 1 1]/3, 1, x(:));
out(1:2) = [x(1);sum(x(1:2))/2];

Weitere Antworten (1)

Image Analyst
Image Analyst am 22 Mär. 2013
Bearbeitet: Image Analyst am 22 Mär. 2013
To get the middle chunk, do this:
yMiddle = conv(x, [1 1 1]/3, 'valid');
Then prepend and append the endpoints as you already calculated.
Edit: Haven't heard from you, so here's a full-blown demo for you. It shrinks the window as it gets closer to the edges. Let me know if you need a version robust enough to handle any window size.
x = 1:10
% Get the first and last elements.
yFirst = mean(x(1:2))
yLast = mean(x(end-1:end))
% Now get the middle with a moving average.
yMiddle = conv(x, [1 1 1]/3, 'valid')
yFinal = [yFirst yMiddle, yLast]
In the command window you see:
x =
1 2 3 4 5 6 7 8 9 10
yFirst =
1.5000
yLast =
9.5000
yMiddle =
2.0000 3.0000 4.0000 5.0000 6.0000 7.0000 8.0000 9.0000
yFinal =
1.5000 2.0000 3.0000 4.0000 5.0000 6.0000 7.0000 8.0000 9.0000 9.5000

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by