home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.next.programmer
- Path: sparky!uunet!ukma!eng.ufl.edu!news
- From: statman@stat.ufl.edu (charles d. kincaid)
- Subject: DSP questions
- Message-ID: <1993Jan27.221053.570@eng.ufl.edu>
- Sender: news@eng.ufl.edu (Usenet Diskhog System)
- Reply-To: statman@stat.ufl.edu
- Date: Wed, 27 Jan 93 22:10:53 GMT
- Lines: 26
-
- Hello,
- I am trying to do a convolution between a data array and a filter.
- I am looking into using the DSP for this (hey, it's there! why not!).
- However, I am having some problems getting into the right mindset. If
- some very kind person can help I would appreciate it very much. This is
- a small part of my work for my dissertation and it's bogging me down.
-
- First, the DL documentation says that when two vectors are multiplied
- the range of the data must be between -1 and 1. Does it matter how this
- is done? Can I just subtract the minimum and then divide by the range to
- get into -1 to 1? Then after the fft do I just multiply and add to get
- the right answer?
- I am using the fft.c code in the DSP folder of the Examples. It
- does an FFT, but not an inverse FFT. Can the inverse FFT be calculated
- using DSPAPfftr2a?
- As you can see, I am new to Fourier transforms. That may be
- part of my problem. I originally intended to use the Numerical Recipes
- in C book, but their method isn't working properly. Any help is
- greatly appreciated.
-
- Sincerely,
-
- charles d. kincaid
- statman@stat.ufl.edu
- ------------------------------------------------------------------------------
- Dept. of Statistics \_ \_ \_\_\_\_\_ 'Damn fine coffee...and hot, too!'
-