Hello,
I've got question about signal processing flow here.
Why are we using boxes as follows: Simple DSP (x^2), Signal average and Simple DSP (log(1+x)), before classifying?
I'm asking about main purpose - isn't it easier to compute power spectrum and find the highest peak in the given range?
Thanks in advance for response.
Cheers,
Matti.
SSVEP, scenario: ssvep-bci-4-classifier-training
-
- Posts: 112
- Joined: Sun Mar 14, 2010 12:58 pm
Re: SSVEP, scenario: ssvep-bci-4-classifier-training
in the SSVEP scenario, squaring a signal band-pass filtered in a given frequency band and then averaging it over time actually means computing the power of the signal in that band. Alternatively you could indeed also use an FFT to compute the power of the signal in a given frequency bin, this would be more or less equivalent. I don't know which one works best or which one is easier (both are fairly easy ), but both approaches should work. Let us know if you make a comparison!
Best regards,
Fabien
Best regards,
Fabien