Replacing a for loop?

Hi,

I've written a small script to calculate the short time energy function of a signal. However it feels a bit clunky and my data sets are quite large so processing takes a while. Is there a way to speed up this process using vectorisation? This is the code at the moment.

 function EImat=EI_mod(A,seg_length)

 rms=sqrt(((sum(A.^2))/length(A)));

 for i=1:(length(A))/seg_length;

    A1=A((i*seg_length)-(seg_length-1):(i*seg_length));

    rms1=sqrt(((sum(A1.^2))/seg_length));

    EImat(i)=rms1;

 end

Any help will be greatly appreciated.

Thanks

Jack

[dpb]

nSeg=floor(length(A)/seg_length);
EImat=zeros(1,nSeg);
for i=1:nSeg
  ...

Gotta' run--but preallocating will surely at least help...

Answer 1

[Roger Stafford]

EImat=sum(reshape(A,seg_length,[]).^2,1)/seg_length;

(Note: I assume here that the length of A is some integral multiple of seq_length.)

(Note 2: You don't return the 'rms' quantity in your function, so I haven't computed it here.)