Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site faron.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!unc!mcnc!philabs!cmcl2!harvard!bbnccv!bbncca!linus!faron!sdo
From: sdo@faron.UUCP (Sean David O'Neil)
Newsgroups: net.math,net.wanted,net.graphics
Subject: Re: fast convolution?
Message-ID: <388@faron.UUCP>
Date: Mon, 18-Nov-85 13:16:58 EST
Article-I.D.: faron.388
Posted: Mon Nov 18 13:16:58 1985
Date-Received: Wed, 20-Nov-85 20:58:35 EST
References: <1484@Shasta.ARPA> <274@geowhiz.UUCP>
Reply-To: sdo@faron.UUCP (Sean David O'Neil)
Distribution: net.math,net.wanted,net.graphics
Organization: The MITRE Coporation, Bedford, MA
Lines: 27
Xref: watmath net.math:2542 net.wanted:7605 net.graphics:1289
Summary: Better reference

In article <274@geowhiz.UUCP> karsh@geowhiz.UUCP (Bruce Karsh) writes:
>In article <1484@Shasta.ARPA> samadani@Shasta.UUCP (Ramin Samadani) writes:
>>Can anyone suggest articles or books on the most recent techniques for
>>fast convolutions? 
>
> One real good book with a lot of info on fast convolution algorithms is
>called "Number Theory in Digital Signal Processing", by James H. McClellan
>and Charles M. Rader.  It is published by Prentice Hall.
>-- 
>Bruce Karsh

I think this book is a good primer on recent (i.e. up to 1978)
developments in number theoretic DSP.  However, it is definitely
not the best place to find out about the latest and best algortithms
available.  For instance, it came out before the relative speed of
the Winograd FFT had been evaluated.  A much better starting
place, and a much more usable book, is "Fast Fourier Transform
and Convolution Algorithms", by H.J. Nussbaumer, 2nd ed.  It is
published by Springer-Verlag in the Springer Series on Information
Sciences.

____

Sean O'Neil
MITRE Corp.
Mailstop E020
Burlington Rd, Bedford, MA 01730