[Eeglablist] IIR filter cutting slope

[Andreas Widmann]

Hi Magda,

you should get a description of the filter stating cutoff, transition bandwidth, and order (don’t you?), e.g.

> LPF has cutoff of 48.0 Hz, transition bandwidth of 1.0 Hz and its order is 12.0


All IIR filters, including the elliptic filters implemented by pop_iirfilt, do theoretically converge somewhere in the magnitude response to -6 dB / oct per order. In the example above this would result in a theoretical roll-off of -6 * 12 * 2  = -144 dB/oct (times two due to two-pass forward and reverse filtering as specified by your parameter set). In many IIR filters, however, this slope is  might only be reached at rather narrowband or irrelevant frequency ranges far away from the cutoff frequency (in particular also Butterworth filters). For elliptic IIR filters the transition bandwidth is a much better and more reliable (and replicable) description of the roll-off slope (similar as for FIR filters).

Hope this helps! Best,
Andreas

Am 25.06.2014 um 16:13 schrieb Magda Gawłowska <mag.gawlowska at gmail.com>:

> I have a question regarding IIR filter. 
> 
> If I use this command line:
> EEG = pop_iirfilt( EEG, 1, 0, [], 0, 0);
> As well as
> EEG = pop_iirfilt( EEG, 0, 48, [], 0, 0);
> 
> How can I determine the filter's cutting slope in dB/octave??
> 
> 
> Thank you in advance!
> Magda
> 
> 
> _______________________________________________
> 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