[Eeglablist] filter order

Andreas Widmann widmann at uni-leipzig.de
Thu Jun 15 08:33:47 PDT 2017


Hi Jaleh,

every digital FIR or IIR filter has a filter order. Mathematically the order is the delay from input to output. As we filter offline and usually compensate for any delays the one or the other way, the filter order per se is usually not relevant. However, the order also defines the maximum steepness of filter roll-off in the frequency domain. The higher the order the steeper the roll-off between passband and stopband. For FIR filters the transition band width is a more useful concept than roll-off. Transition band with is the frequency difference between passband and stopband and can be computed from order for most common FIR filter types. The higher the order the narrower the transition band. Note, that one cannot directly compare the order of IIR and FIR filters as they are implemented differently. Rather effective impulse response duration should be compared.

There is also a free full-text version of the paper mentioned by Tarik introducing the basic concepts:
http://home.uni-leipzig.de/~biocog/eprints/widmann_a2015jneuroscimeth250_34.pdf

Hope this helps! Best,
Andreas

> Am 12.06.2017 um 08:34 schrieb Jaleh Mohammad alipoor <sugmad973 at yahoo.com>:
> 
> hello. 
> l have filtered my data between 1 to 45 Hz but my professor think l should apply a filter order to the data. l wonder filter order is a thing related to the time scale and filtering is related to frequency. how they are related to each other? and how can we decide what amount of filtering order is suitable? what does filtering order do exactly??
> l would appreciate any help (especially in detail!). 
> 
> many thanks
> Zhaleh
> _______________________________________________
> 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



More information about the eeglablist mailing list