NetNews Usenet Archive 1992 #26/NN_1992_26.iso/spool/comp/dsp/2577

Internet Message Format | 1992-11-13 | 1.7 KB

Path: sparky!uunet!charon.amdahl.com!pacbell.com!sgiblab!zaphod.mps.ohio-state.edu!cis.ohio-state.edu!rutgers!cmcl2!panix!davidc
From: [e-mail] (David B. Chorlian)
Newsgroups: comp.dsp
Subject: Impulse Response Filter Coeffs Needed
Summary: Impulse Response Coeffs needed for 1024 point high pass FIR filter
Keywords: Filter, FIR, IEEE, FFT
Message-ID: <[e-mail]>
Date: 14 Nov 92 01:47:58 GMT
Organization: PANIX Public Access Unix, NYC
Lines: 29


    I am trying to construct a 1024 point high pass FIR filter using
the IEEE Digital Signal Processing program eqfir.f.  I was successful
in constructing 256 point filters, but although I enlarged the array sizes
to accomodate contructing 1024 point filters, the program failed to
converge and then crashed with a segmentation fault.  Not being a FORTRAN
programmer and being in somewhat of a hurry, I appeal to someone on the
net to supply me with the coefficients for a high-pass filter with the
cutoff point at 2Hz (1024pts = 1 sec).  Or, if someone has a better idea
for accomplishing the same goal, I'd appreciate the advice.  The coefficients
are used for digital filtering using an fft of the signal, an fft of the
coefficients, their convolution (since the coefficients are symmetric around
the center of the array, a simple term by term multiplication is fine), and
an inverse fft to get the filtered signal.  If you have or can easily
generate the coefficients, please email them to me.  If you have other ideas,
use your judgment on posting.
This project involves the processing of EEG data.
If you want to know more about it, email and I will give you the details.

David B. Chorlian
Neurodynamics Lab
SUNY/HSCB
[e-mail]

--
David B. Chorlian
Neurodynamics Lab
SUNY/HSCB
[e-mail]