[Eeglablist] firwsord dB question

[Andreas Widmann]

Hi Russell,

> Should devdb be calculated as 10 * log10 or as is, 20 * log10?
20. Ripple and attenuation commonly refers to amplitude not power in filter design. You may lookup the equations in many textbooks. If you have MATLAB signal processing toolbox you find them for example on the kaiserord doc-page.

> The value of dev (.0002 for Kaiser window) describes the desired ripple limit as a ratio of desired to original signal, both in microVolts I think.
dev/ripple/attenuation is the maximum allowed deviation of the frequency response from the requested frequency response (1 in passband, 0 in stopband). You may therefore compute the minimum stopband attentuation as 20 * log10( dev ), here -74 dB.

> However, as we don't necessarily want the power, should decibels be calculated as a regular log ratio (10 * log10)?
Not sure what you mean by "regular", When dealing with amplitude you typically use 20 * log10 and and when dealing with power you typically use 10 * log10. Using it this way you get the same numbers for amplitude and power what is very convenient in many situations:
20 * log10( x ) = 10 * log10( x .^ 2 )

Hope this helps! Best,
Andreas

> I frequently get confused with voltages and power so I could very well be, and likely am, wrong. Just wanted to verify the function was correct as written. Thanks very much for your help and teaching!
> 
> Very respectfully,
> 
> Russell T. Toll, Ph.D.
> Assistant Professor
> Department of Psychiatry
> Center for Depression Research and Clinical Care
> UT Southwestern Medical Center
> 6363 Forest Park Rd, Suite BL 13.206
> Dallas, TX 75235-5479
> 
> 
> ________________________________
> 
> UT Southwestern
> 
> 
> Medical Center
> 
> 
> 
> The future of medicine, today.
> 
> _______________________________________________
> 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
>