[Eeglablist] Filtering issues

[Andreas Widmann]

Hi,

googling the list archives (http://www.google.com/search?q=highpass+site:sccn.ucsd.edu/pipermail/eeglablist/) gives you several threads discussing this problem, e.g.
http://sccn.ucsd.edu/pipermail/eeglablist/2008/002210.html

In a nutshell: The filter order as suggested by EEGLAB is too high for this extreme cutoff-frequency. Use higher cutoff or different filter type. Transition band width is defined as cutoff * 0.15 for firls-based filters in EEGLAB IIRC.

Note:
I recommend upgrading to a more recent version of EEGLAB where fir1 is used as default filter function (in case you did not select firls explicitly). firls-based filtering does not give optimal results (http://www.frontiersin.org/Perception_Science/10.3389/fpsyg.2012.00233/full).

Best,
Andreas

Am 21.09.2012 um 17:03 schrieb tmle at notes.cc.sunysb.edu:

> Hi eeglablist,
> 
> I have been attempting to apply a highpass filter to our 64-channel data set. So far I have only resampled (to 256) and rereferenced (to average reference) prior to this step. However, when I enter .01 for the lower edge in the basic filter option, eeglab will run for about 20 minutes then give me an out of memory error (eeglab error in function firls() at line 150). Could you please let me know why this might be the case? I'm not sure whether my laptop memory is a problem as it has 8G of RAM. Also, eeglab says the eegfilt () - highpass transition band width is 0.001Hz. Shouldn't this be .01 like I have entered for the lower edge?
> 
> Thanks,
> 
> Le
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu