https://wiki.hh.se/ceres/index.php?action=history&feed=atom&title=Publications%3ARelating_FFTW_and_Split-RadixPublications:Relating FFTW and Split-Radix - Revision history2026-05-22T17:28:26ZRevision history for this page on the wikiMediaWiki 1.22.6https://wiki.hh.se/ceres/index.php?title=Publications:Relating_FFTW_and_Split-Radix&diff=4275&oldid=prevSlawek: Created page with "<div style='display: none'> == Do not edit this section == </div> {{PublicationSetupTemplate|Author=Oleg Kiselyov, Walid Taha |PID=588272 |Name=Kiselyov, Oleg (Monterey, CA 93..."2014-06-26T03:44:55Z<p>Created page with "<div style='display: none'> == Do not edit this section == </div> {{PublicationSetupTemplate|Author=Oleg Kiselyov, Walid Taha |PID=588272 |Name=Kiselyov, Oleg (Monterey, CA 93..."</p>
<p><b>New page</b></p><div><div style='display: none'><br />
== Do not edit this section ==<br />
</div><br />
{{PublicationSetupTemplate|Author=Oleg Kiselyov, Walid Taha<br />
|PID=588272<br />
|Name=Kiselyov, Oleg (Monterey, CA 93943);Taha, Walid [waltah] (Department of Computer Science, Rice University)<br />
|Title=Relating FFTW and Split-Radix<br />
|PublicationType=Conference Paper<br />
|ContentType=Refereegranskat<br />
|Language=eng<br />
|Journal=<br />
|JournalISSN=<br />
|Status=<br />
|Volume=<br />
|Issue=<br />
|HostPublication=<br />
|Conference=ICESS'04 International Conference on Embedded Software and Systems<br />
|StartPage=<br />
|EndPage=<br />
|Year=2004<br />
|Edition=<br />
|Pages=<br />
|City=<br />
|Publisher=<br />
|Series=<br />
|SeriesISSN=<br />
|ISBN=<br />
|Urls=<br />
|ISRN=<br />
|DOI=<br />
|ISI=<br />
|PMID=<br />
|ScopusId=<br />
|NBN=urn:nbn:se:hh:diva-20974<br />
|LocalId=<br />
|ArchiveNumber=<br />
|Keywords=<br />
|Categories=Datorsystem (20206)<br />
|ResearchSubjects=<br />
|Projects=<br />
|Notes=<br />
|Abstract=<p>Recent work showed that staging and abstract interpretation can beused to derive correct families of combinatorial circuits, and illustrated this technique with an in-depth analysis of the Fast Fourier Transform (FFT) for sizes 2n.While the quality of the generated code was promising, it used more floatingpoint operations than the well-known FFTW codelets and split-radix algorithm.This paper shows that staging and abstract interpretation can in fact be used toproduce circuits with the same number of floating-point operations as each ofsplit-radix and FFTW. In addition, choosing between two standard implementations of complex multiplication produces results that match each of the twoalgorithms. Thus, we provide a constructive method for deriving the two distinctalgorithms.</p><br />
|Opponents=<br />
|Supervisors=<br />
|Examiners=<br />
|Patent=<br />
|ThesisLevel=<br />
|Credits=<br />
|Programme=<br />
|Subject=<br />
|Uppsok=<br />
|DefencePlace=<br />
|DefenceLanguage=<br />
|DefenceDate=<br />
|CreatedDate=2013-01-14<br />
|PublicationDate=2013-01-15<br />
|LastUpdated=2014-01-24<br />
|diva=http://hh.diva-portal.org/smash/record.jsf?searchId=1&pid=diva2:588272}}<br />
<div style='display: none'><br />
<br />
== Keep all hand-made modifications below ==<br />
</div>{{PublicationDisplayTemplate}}</div>Slawek