home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: alt.lang.basic
- Path: sparky!uunet!caen!destroyer!news.iastate.edu!kwj
- From: kwj@iastate.edu (kwj)
- Subject: Re: Discrete FFT in basic
- Message-ID: <BxtIB1.3n4@news.iastate.edu>
- Sender: news@news.iastate.edu (USENET News System)
- Organization: Iowa State University, Ames, IA
- References: <BxsM4w.EM3@ns1.nodak.edu>
- Distribution: na
- Date: Mon, 16 Nov 1992 16:40:10 GMT
- Lines: 29
-
-
- An *excellent* book for numerical analyses is
-
- Press, W.H., Flannery, B.P., Teukolsky, S.A., and Vetterling,W.T.
- (1989). Numerical Recipes: The Art of Scientific Computing,
- Cambridge Unversity Press.
-
- I use the FORTRAN version, which can easily be translated into BASIC. I
- believe that a BASIC version has been published, but I have not looked at
- it yet.
-
- Note: Since the FFT is a convoluted (pun intended) algorithm, many
- algorithms have errors in them, including the one in Numerical
- Recipes. If you use the FORTRAN version, the algorithm has the
- line
-
- WPI = DSIN(THETA)
-
- This should be
-
- WPI = -DSIN(THETA)
-
- This error will give you problems if you have a complex data set.
-
- I will post an algorithm that I wrote for a real (not complex) data
- set. The algorithm is based on that by Press et al.
-
- kwj
-
-