help-octave
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

troubles with remez


From: Dan McMahill
Subject: troubles with remez
Date: Mon, 21 Jun 2010 09:21:11 -0400
User-agent: Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.1.10) Gecko/20100512 Thunderbird/3.0.5

Hello,

I'm using octave-3.2.4 and signal-1.0.10. I'm having problems getting the remez() function to work and give reasonable answers.

The filter I'm trying to do should be fairly simple. It is just something which has a prescribed amplitude vs freq rolloff over about 15% of the Nyquist band. I'm only asking for a fairly low order (less than 10) filter. It seems I can easily get "error: remez: insufficient extremals" and also fairly easily get "warning: remez: --failed to converge -- returned filter may be bad.".

The same code runs fine under matlab.

For example

b = remez(7, ...
   [0.0  0.24 0.25 0.35 0.36 1.0], ...
   [1.0  1.0  1.0  0.7  0.7  0.0], ...
   [0.01      1.0       0.01])

gives the too many extremals where matlab gives a filter which matches the desired slope over the 0.25 to 0.35 band.

b = remez(6, ...
   [0.0  0.24 0.25 0.35 0.36 1.0], ...
   [1.0  1.0  1.0  0.7  0.7  0.0], ...
   [0.01      1.0       0.01])

gives the failed to converge.

on matlab, I also get a filter which matches the requested response over the 0.25 to 0.35 band.

am I doing something wrong or is the octave remez() just in need of a little work?

Thanks
-Dan


reply via email to

[Prev in Thread] Current Thread [Next in Thread]