Sign in

username:

password:



Not a member?

Search audiodsp



Search tips

Subscribe to audiodsp



audiodsp by Keywords

AAC | ADPCM | Convolution | DAFx | FFT | IIR | Mixer | MP3 | MPEG | MPEG-4

Ads

Discussion Groups

Discussion Groups | Audio Signal Processing | Automatic filter coefficient generation from given freq?

Technical discussions related to Audio Signal Processing (digital effects, acoustics, noise reduction, musical signal processing, etc).

  

Post a new Thread

Automatic filter coefficient generation from given freq? - g_lo...@yahoo.co.uk - Mar 12 8:28:17 2007



Hi.
I'm working on a real-time automatic acoustic feedback elimination system, much the same as
other attempts documented on the web. It will fft the input signal, detect the peak
corresponding to the characteristic spike in the spectrum, and filter the detected frequency
with a notch/band-stop filter. I haven't decided on the exact filter I will use yet, I'm
currently experimenting with FIRs in MATLAB.

My question is: Can the appropriate coefficients be generated for the filter from just a
knowledge of the frequency value to be filtered? I'm imagining an algorithm which takes the
result of the peak detection as in put, and outputs the desired coefficients. 

This vital part of the process doesn't seem to be described in any of the feedback elimination
papers I've found. I would be very greatful if anyone could even point me in the right
direction, as I'm a bit of a green horn in the filter design area. 

Thanks,
Ger.



(You need to be a member of audiodsp -- send a blank email to audiodsp-subscribe@yahoogroups.com )

Re: Automatic filter coefficient generation from given freq? - Amaresh patil - Mar 13 7:58:08 2007

Hi
  You can raplace the component with tunable Notch filter. This is well explained in the S.K
mitra book.

  All the best
  Amaresh
  
g...@yahoo.co.uk wrote:
          Hi.
I'm working on a real-time automatic acoustic feedback elimination system, much the same as
other attempts documented on the web. It will fft the input signal, detect the peak
corresponding to the characteristic spike in the spectrum, and filter the detected frequency
with a notch/band-stop filter. I haven't decided on the exact filter I will use yet, I'm
currently experimenting with FIRs in MATLAB.

My question is: Can the appropriate coefficients be generated for the filter from just a
knowledge of the frequency value to be filtered? I'm imagining an algorithm which takes the
result of the peak detection as in put, and outputs the desired coefficients. 

This vital part of the process doesn't seem to be described in any of the feedback elimination
papers I've found. I would be very greatful if anyone could even point me in the right
direction, as I'm a bit of a green horn in the filter design area. 

Thanks,
Ger.



(You need to be a member of audiodsp -- send a blank email to audiodsp-subscribe@yahoogroups.com )