Matrix inversion using DSP 56K

Article View: comp.dsp

Article #833

#833

From: Aviv Siegel
Date: Wed, 13 Jun 1990 16:34

21 lines
1665 bytes

        Does anyone have a matrix-inversion program for the dsp56000 (that
works)? I have been working on it for a while and there seems to be a
problem implementing it with fractions, since either the original or the
inverted matrix have coefficients larger than one.
        I know this is not a usual DSP problem - a high-language solution
on a microprocessor is probably easier. But the decision was made by
someone else and I was left to implement it...
        I tried a C program that worked OK on PC but when compiled (by
motorola cc56000 compiler) and run on a Loughborough card it failed
claiming that "the program damaged the user interface".
        I prefer am assembly program but any solution is welcome.

        please email me directly if possible to EESADEE$VMSA.TECHNION.AC.IL

/**************************************************************************
     --     ---\  -----  ----- * (And what use is life if we can't complain
    /  \    |   \ |      |     * about it?)
   /----\   |   | |----  |---- ********************************************
  /      \  |   / |      |     * Adee Ran (EESADEE@vmsa.technion.ac.il)
 /        \ ---/  -----  ----- *******************************************/

Technical Details

Message-ID: <9719@discus.technion.ac.il>

Path: arc1-ca.pugleaf.net!utzoo!attcan!uunet!tut.cis.ohio-state.edu!ucbvax!techunix.BITNET!aviv

Show Full Headers

Complete Article Headers

Path: arc1-ca.pugleaf.net!utzoo!attcan!uunet!tut.cis.ohio-state.edu!ucbvax!techunix.BITNET!aviv
From: aviv@techunix.BITNET (Aviv Siegel)
Newsgroups: comp.dsp
Subject: Matrix inversion using DSP 56K
Message-ID: <9719@discus.technion.ac.il>
Date: 13 Jun 90 16:34:07 GMT
Sender: daemon@ucbvax.BERKELEY.EDU
Reply-To: aviv%techunix.bitnet@jade.berkeley.edu (Aviv Siegel)
Organization: Technion, Israel Inst. Tech., Haifa Israel
Lines: 21