[Scilab-users] "Re-sampling" data and making 2 vectors

[Dang, Christophe]

Hello Jacqueline, 

> I have to "cut" the matrixes, they should have the same lenght (because
> I need Data2-(Data1-mean(Data1)) , and starts at the comun hour. I dont
> know how do it, Any suggest?

Do you have a problem with the resampling algorithm or with the data
handling?

Concerning the algorithm, I suggest you perform a kind of Savitzky-Golay
smoothing of the biggest matrix :

1- define a, say 3rd degree polynomial by regression on a sliding of 5 
points (must be an odd number),

2- keep the value of the polynomial corresponding to the data in the 1st
matrix that lay in the sliding window.

You cannot use the Savitzky-Golay convolution coefficients, as you are
not necessarilly interested in the middle value.

You can have an example of implementation here
http://commons.wikimedia.org/wiki/File:Savitzky-golay_pic_gaussien_bruite.svg
(look at the last solution; the first one uses the convolution
coefficients).

Hope this helps.

-- 
Christophe Dang Ngoc Chan
Mechanical calculation engineer
This e-mail may contain confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error), please notify the sender immediately and destroy this e-mail. Any unauthorized copying, disclosure or distribution of the material in this e-mail is strictly forbidden.