nemu
66 lines
2.7 KiB
Plaintext
66 lines
2.7 KiB
Plaintext
/**************************************************************************
|
|
* Parks-McClellan algorithm for FIR filter design (C version)
|
|
*-------------------------------------------------
|
|
* Copyright (c) 1995,1998 Jake Janovetz (janovetz@uiuc.edu)
|
|
*
|
|
* This library is free software; you can redistribute it and/or
|
|
* modify it under the terms of the GNU Library General Public
|
|
* License as published by the Free Software Foundation; either
|
|
* version 2 of the License, or (at your option) any later version.
|
|
*
|
|
* This library is distributed in the hope that it will be useful,
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
|
* Library General Public License for more details.
|
|
*
|
|
* You should have received a copy of the GNU Library General Public
|
|
* License along with this library; if not, write to the Free
|
|
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
|
|
*
|
|
*************************************************************************/
|
|
|
|
DESCRIPTION:
|
|
============
|
|
The Parks-McClellan program is a filter design program which creates
|
|
optimal filters in the sense that there is equirriple in the pass bands
|
|
and stop bands of the resulting frequency response.
|
|
|
|
This version was written for use in another program in which I required
|
|
C access to the design routine. Therefore, Matlab's 'remez' function
|
|
wouldn't do. I also didn't like all the FORTRAN to C conversions done
|
|
because they're impossible to read. This version is based on the original
|
|
FORTRAN code and equations presented in Oppenheim & Schafer.
|
|
|
|
To use the code, simply include all but 'main' in your own code. There
|
|
is no front end supplied with this code, but that should be simple to
|
|
implement.
|
|
|
|
If any errors are found, please let me know. I have compared the output
|
|
to that of Matlab's 'remez' function and get close (within reasonable
|
|
rounding error -- on the order of 1e-13) to the impulse response. However,
|
|
I have not exhaustively tested it with a lot of vectors.
|
|
|
|
Jake
|
|
|
|
COMPILE INSTRUCTIONS:
|
|
=====================
|
|
To compile the test program that generates a single filter, but asks for
|
|
no input (it truly is a test program), very simply type:
|
|
|
|
cc -o test test.c remez.c -lm
|
|
|
|
On a UNIX machine. You may use other compilers (namely gcc), but you
|
|
must link the math libraries (-lm). If you have problems compiling on
|
|
a system, let me know and I'll see what I can do. The code should be
|
|
very portable, though.
|
|
|
|
Then, you can run the compiled program (test) and watch it generate some
|
|
fascinating (!) coefficients.
|
|
|
|
THANKS:
|
|
=======
|
|
Thanks to Dr Peter Kootsookos for his help in finding a couple bugs in
|
|
the original code.
|
|
|
|
|