Optimal Wavelength Converter Placement in Optical Networks by Genetic Algorithm

Search this article

Abstract

In optical networks, wavelength converters are required to improve the efficiency of wavelength-division multiplexing. In this paper, we propose a genetic algorithm to determine the optimal locations of the nodes in the network where a given number of converters are placed. Optimality is achieved by the minimum wavelength blocking probability. Our algorithm is applied to two realistic networks constructed from the locations of major cities in Ibaraki Prefecture and from those in Kanto District in Japan and is shown to reach the nearly optimal solution in a limited number of generations. The accuracy is verified by simulation. The computational time is compared with that of an exhaustive search algorithm.

Journal

References(13)*help

See more

Details

  • CRID
    1570009752558942848
  • NII Article ID
    110003219435
  • NII Book ID
    AA10826261
  • ISSN
    09168516
  • Text Lang
    en
  • Data Source
    • CiNii Articles

Report a problem

Back to top