How to compute power spectral density (PSD) using FFT for each channel of the device and calculate average power of each band?

I'm entirely new to Matlab and EEGLab. I need to compute power spectral desnity (PSD) using FFT for each channel of the device that I'm using, to classify the brainwaves generated from each channel and calculate the average relative power of each band. Please help.

[dpb]

See example in documentation <fft>. As noted if you arrange your data where each channel is a column, then FFT will operate and return the FFT for each channel directly; you can get the PSDs for each by continuing the same pattern on the output arrays as is shown in the example for one channel.

Just give it a go and let us know where you run into a particular issue...

[Surakshya Pradhan]

Hi,

I have a .set file and I'm not sure how to arrange so each channel is a column. Also, how to classify waves as alpha, beta, theta etc. from each channel?

[dpb]

If the file is sequential as would presume it would be, that's what you'll get automagically.

As far as the type of waves, that's a domain of application question, not one about MATLAB. I'm a NucE, so "I know nuthink!" about EEG specifically; I'm sure the literature is full of analysis techniques.

[Surakshya Pradhan]

Thank you so much for your help!