EEGLAB workflow
- 1) read in data
- 2) change sampling rate to 256 (this makes 154 points between a [-0.3 0.3] epoch)
- 3) parse out the right hand and left hand epochs
- 4) take the average over all 64 electrodes (next try only working with nodes in the 'mu area')
- 5) take the fft of the average
- 6)parse the epochs
3 comments:
You must be careful when using the FFT for analysis of spatial/temporal sensitive data. In your case, taken the FFT over the full data would not suffice. As the FFT has no temporal localization, which means that it will provide you spectra information over the full range of the series. You must also be careful with end points as they can cause problems in the frequency domain.
You may want to consider a spectra that has temporal localization like Wavelet xforms.
Side note: you can actually build your ANN to do FFT and Wavelets too.
Thanks for your comments. I actually took the FFT of the individual event columns. Thanks for clarifying why!
I wonder if I could just take a few more miliseconds of data like from [-0.4 0.1] per event time frame but only use the FFT data from [-0.3 0.05]?
I'll look into 'wavelet xforms' tomorrow and Friday.
Thanks again
And I'll also look into Windowed-Fourier analysis
Post a Comment