Re: Maxima locating in noisy data

[Sergei Steshenko]

--- On Tue, 6/29/10, dirac <address@hidden> wrote:

> From: dirac <address@hidden>
> Subject: Maxima locating in noisy data
> To: address@hidden
> Date: Tuesday, June 29, 2010, 2:30 AM
> 
> Hi again,
> 
> I am unsure if this is the purpose of the forum, but I am
> having problems
> finding a way to locate peaks in some noisy data. It is
> basically a
> Fabry-Perot laser spectrum which is stored in a matrix of
> 10002 lines and 21
> columns...This is terribly vague I know but any suggestions
> would be awesome
> 
> Thanks in advance
> Martin
> 
> -----
> Converting myself from Excel to Octave for reasons I hope
> to find out.
> -- 
> View this message in context: 
> http://octave.1599824.n4.nabble.com/Maxima-locating-in-noisy-data-tp2271936p2271936.html
> Sent from the Octave - General mailing list archive at
> Nabble.com.
> _______________________________________________
> Help-octave mailing list
> address@hidden
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave
> 

Use a low-pass filter of your liking (e.g. Butterworth) to filter the spectrum 
- this will eliminate narrow peaks.

A widely used approach of gliding average is also application of a low
pass filter.

...

On a side note, since you are dealing with spectrum, have a look at

http://www.unipress.waw.pl/fityk/

- not related to GNU Octave, but maybe a good fit for your problem.

Regards,
  Sergei.