comp.lang.idl-pvwave archive
Messages from Usenet group comp.lang.idl-pvwave, compiled by Paulo Penteado

Home » Public Forums » archive » spectral estimation e.g., MEM, MUSIC
Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend 
Switch to threaded view of this topic Create a new topic Submit Reply
spectral estimation e.g., MEM, MUSIC [message #10701] Wed, 14 January 1998 00:00 Go to next message
Jack Harlan is currently offline  Jack Harlan
Messages: 4
Registered: October 1996
Junior Member
I've checked all the large archives of IDL routines, but haven't been
able to find any existing IDL ports of various modern spectral
estimation techniques aka MSATs. These generally include maximum
entropy methods (such as Burg's algorithm), MUSIC algorithms, etc.
If no one else has done this, I'll probably convert some old
Fortran code myself.
But, if anyone knows of existing IDL code, I'd appreciate hearing
about it.

thanks
jack
--
* Jack Harlan
* NOAA Environmental Technology Laboratory
* 325 Broadway * Boulder, CO 80303
* 303-497-6032 PH * 303-497-3577 FAX
Re: spectral estimation e.g., MEM, MUSIC [message #10777 is a reply to message #10701] Thu, 15 January 1998 00:00 Go to previous message
Jeff Kommers is currently offline  Jeff Kommers
Messages: 2
Registered: June 1997
Junior Member
There is a text by S. Lawrence Marple Jr. called "Digital Spectral
Analysis with Applications". It comes with some Fortran codes that
implement the various spectral analysis techniques, including Burg's
algorithm and MUSIC.

I have written a set of IDL and C wrappers for these Fortran codes so
that I can call them from IDL using call_external(). If you are
interested, post or send me e-mail.

I must warn that using the Fortran codes on different platforms
produced very different results using the test data supplied by
Marple. It appears to be a dependence on floating point
implementation: I got different results using a VAX, Sun (single
precision), and Sun (double precision). The spectra are qualitatively
similar, but the actual numbers differ by as much as 10 percent using
the modified covariance estimator (MODCOVAR). (!)

Jeff
  Switch to threaded view of this topic Create a new topic Submit Reply
Previous Topic: Re: Widget Hierarchy
Next Topic: Help Needed

-=] Back to Top [=-
[ Syndicate this forum (XML) ] [ RSS ] [ PDF ]

Current Time: Wed Oct 08 18:14:09 PDT 2025

Total time taken to generate the page: 0.00579 seconds