Prev: Audio Fingerprinting Technology
Next: skeleton Hilbert transform (emd, hht)
From: vasu on 9 Feb 2006 07:14
we are currently working on echo cancellation using adaptive filtering on
dsk 6713. we have implemented fir low pass and high pass filters on the
board using CCS.
we have developed L.M.S algorithm in Matlab(Not the Blockset provided in
Help. A new one).
we want to download that m file on the board(Portability between Matlab
and Code Composer Studio).
How ?


From: mlimber on 9 Feb 2006 09:00
vasu wrote:
> we are currently working on echo cancellation using adaptive filtering on
> dsk 6713. we have implemented fir low pass and high pass filters on the
> board using CCS.
> we have developed L.M.S algorithm in Matlab(Not the Blockset provided in
> Help. A new one).
> we want to download that m file on the board(Portability between Matlab
> and Code Composer Studio).
> How ?

You will likely need to convert the MATLAB code to C or C++ since
MATLAB doesn't run on DSPs. There may be some automated tools that can
do this for you (I could be wrong, but I think the MathWorks'
conversion tools only work for platforms already supported by MATLAB
since they require you link their libraries in). I'd suggest you ask in
a MATLAB newsgroup for help doing this.

Cheers! --M

From: goldfita on 9 Feb 2006 23:42

vasu wrote:
> we are currently working on echo cancellation using adaptive filtering on
> dsk 6713. we have implemented fir low pass and high pass filters on the
> board using CCS.
> we have developed L.M.S algorithm in Matlab(Not the Blockset provided in
> Help. A new one).
> we want to download that m file on the board(Portability between Matlab
> and Code Composer Studio).
> How ?


I have faced the same problem. I believe there is a (expensive) matlab
tool that will target c67xx, but I'm not sure. And there are issues.
Has the code been written with real-time issues in mind? I suggest you
just convert to C. It should only take a few days work. Use a mex
file to develop the real-time code in matlab first. Then port it to
the dsp. This should be very easy if you think about what you're doing
before you start coding. There may be some sample code on my home
page.

--
-Todd
http://www.signalsguru.net/

 | 
Pages: 1
Prev: Audio Fingerprinting Technology
Next: skeleton Hilbert transform (emd, hht)