Optimization of time-domain header processing filters using genetic algorithms

Mihaela Dinu, Inuk Kang, Mahmoud Rasras, James J. Jaques

Research output: Contribution to journalArticle

Abstract

We present genetic algorithms as optimization tools for the design of header recognition filters based on time-delay lines. The fitness parameter in the selection process is the discrimination of the target against all other bit patterns. For phase-shift-keyed headers, optimized filters exhibit improved discrimination compared with the matched filter. Time-domain filtering with no ambiguities is also achieved for specific amplitude-shift-keyed headers. Delay-line-based time-domain filters can be implemented in a planar lightwave circuit platform by simple phase and amplitude adjustments on the time delay taps.

Original languageEnglish (US)
Pages (from-to)297-299
Number of pages3
JournalOptics Letters
Volume33
Issue number3
DOIs
StatePublished - Feb 1 2008

    Fingerprint

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics

Cite this