Pattern recognition in cognitive communication

Ozgur Orcay*, Berk Ustundag

*Corresponding author for this work

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

11 Citations (Scopus)

Abstract

A cognitive communication method based on recognition of continuous signal patterns respecting to digital data streams is proposed in this study. Digital signals to be transferred are first converted to optimally chosen patterns from a predefined glossary with respect to bandwidth requirement and medium conditions. Receivers recognize the patterns from the chosen frequency band of the spectrum via neural network. The received patterns are matched to respecting data sequence within the common glossary to recover the information. The purpose of this study is increasing the spectrum efficiency via manageable Signal to Noise Ratio. Pattern based real-time encoding and neural network based decoding technique is used here. It is shown that the transmitted information by multiple sources can be decoded by the receivers in the same time interval, frequency band and location by using this method.

Original languageEnglish
Title of host publication2008 23rd International Symposium on Computer and Information Sciences, ISCIS 2008
DOIs
Publication statusPublished - 2008
Event2008 23rd International Symposium on Computer and Information Sciences, ISCIS 2008 - Istanbul, Turkey
Duration: 27 Oct 200829 Oct 2008

Publication series

Name2008 23rd International Symposium on Computer and Information Sciences, ISCIS 2008

Conference

Conference2008 23rd International Symposium on Computer and Information Sciences, ISCIS 2008
Country/TerritoryTurkey
CityIstanbul
Period27/10/0829/10/08

Keywords

  • Artificial neural network
  • Cognitive communication
  • Noise immunity
  • Pattern recognition
  • Spectrum management

Fingerprint

Dive into the research topics of 'Pattern recognition in cognitive communication'. Together they form a unique fingerprint.

Cite this