[Eeglablist] Dealing with data of the same sample from different channels

[Sahar Selim]

Hello,

I am working on 3 channels of the bci competition IV 2a dataset (Fz, Cz, Pz)
I bandpassed each signal, then normalized them.
Now, I have a matrix 3 (channels) x 288 (number of samples) x 750 (number
of points)

I want to calculate the common spatial pattern (CSP) of these signals
I was wondering if I'll be working on data of each channel separately or
would the 3 channels be merged by any mean?

(i.e. the data(1,1,:), data(2,1,:) & data(3,1,:) corresponds to the same
sample but from different electrodes

How would I deal with data that comes from different channels for the same
motor imagery action ex: sample#1 right hand movement ?

Would the feature extraction process of the same sample be done separately
from the data of the same sample the comes from other electrodes?

Thanks in advance
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20150708/d9afaf8f/attachment.html>