http://www.mathworks.com/matlabcentral/newsreader/view_thread/324944
MATLAB Central Newsreader  Stuck in Whitening (Sir Wayne)
Feed for thread: Stuck in Whitening (Sir Wayne)
enus
©19942014 by MathWorks, Inc.
webmaster@mathworks.com
MATLAB Central Newsreader
http://blogs.law.harvard.edu/tech/rss
60
MathWorks
http://www.mathworks.com/images/membrane_icon.gif

Tue, 04 Dec 2012 06:47:08 +0000
Stuck in Whitening (Sir Wayne)
http://www.mathworks.com/matlabcentral/newsreader/view_thread/324944#893024
Md Mahmudul Hasan
Hello everyone and Sir Wayne<br>
<br>
I think i am missing something here. <br>
I am trying to perform a whitening on a sequence (of 1x128), that should reduce the correlation of the sequence. So, as theory says<br>
<br>
[U,D] = eig(cov(X));<br>
Y = sqrt(inv(D)) * U'* X;<br>
<br>
Where X is the sequence and Y should be the Whitened data <br>
<br>
Now here is the problem, if X is a matrix like 128x128 then U and D would be of same dimension, otherwise, in my case, when X is a 1x128 vector, cov(X) returns only the variance of X. and Y doesn't return anything. <br>
<br>
So my question is how to whiten a vector with such dimension to reduce autocorrelation of that sequence. <br>
<br>
Regards<br>
Mahmud