On the use of fuzzy logic for inherently parallel computations

M. Skander Hannachi*, Dong Fangyan, Yutaka Hatakeyama, Kaoru Hirota

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

4 Citations (Scopus)

Abstract

The possibility of using fuzzy logic for performing parallel computation is introduced. An analogy is drawn between non-deterministic models of computing and fuzzy computing. In practice this can be implemented using strings of fuzzy bits, which can represent a "superposition" of several crisp bit strings. Thus, using fuzzy logic and analog fuzzy hardware, inherently single instruction multiple data parallel computing can be achieved. The possibility of speeding up computations using oracle machines using the proposed concept is examined. As a proof of concept, the rapid solution to Deutsch's problem is presented.

Original languageEnglish
Title of host publicationISCIII'07
Subtitle of host publication3rd International Symposium on Computational Intelligence and Intelligent Informatics; Proceedings
Pages89-92
Number of pages4
DOIs
Publication statusPublished - 2007
Externally publishedYes
EventISCIII'07: 3rd International Symposium on Computational Intelligence and Intelligent Informatics - Agadir, Morocco
Duration: 28 Mar 200730 Mar 2007

Publication series

NameISCIII'07: 3rd International Symposium on Computational Intelligence and Intelligent Informatics; Proceedings

Conference

ConferenceISCIII'07: 3rd International Symposium on Computational Intelligence and Intelligent Informatics
Country/TerritoryMorocco
CityAgadir
Period28/03/0730/03/07

Keywords

  • Deutcsh's problem
  • Fuzzy logic
  • Parallel computing
  • Single instructions multiple data
  • Superposition of bit strings

Fingerprint

Dive into the research topics of 'On the use of fuzzy logic for inherently parallel computations'. Together they form a unique fingerprint.

Cite this