• DocumentCode
    1131831
  • Title

    Adaptive beamforming using a novel numerical optimisation algorithm

  • Author

    Roshanaei, Mahnaz ; Lucas, Craig ; Mehrabian, A.R.

  • Author_Institution
    Sch. of Electr. & Comput. Eng., Univ. of Tehran, Tehran, Iran
  • Volume
    3
  • Issue
    5
  • fYear
    2009
  • fDate
    8/1/2009 12:00:00 AM
  • Firstpage
    765
  • Lastpage
    773
  • Abstract
    Currently, the uniform linear array (ULA) is the most commonly used antenna system for different wireless systems like commercial cellular systems. In this study, a ULA adaptive antenna that uses a novel numerical stochastic optimisation algorithm inspired from colonising weeds, designated as invasive weed optimisation (IWO), is introduced. Weeds are shown to be very robust and adaptive to changes in the environment. Thus, capturing their properties leads to a powerful optimisation algorithm. This optimisation algorithm is used for adaptive beamforming; the obtained results are compared with the results obtained from two other optimisation algorithms, that is, the least mean square and genetic algorithms. The reported results show that the IWO is very robust and effective in locating the optimal solution with higher precision and a lower cost function when compared with the other two algorithms. Other advantages of the IWO algorithm is its simplicity and fast convergence, which makes it a practical algorithm for adaptive beamforming.
  • Keywords
    adaptive antenna arrays; array signal processing; genetic algorithms; least mean squares methods; linear antenna arrays; radiocommunication; stochastic processes; adaptive beamforming; antenna system; genetic algorithm; invasive weed optimisation; least mean square; numerical optimisation algorithm; numerical stochastic optimisation algorithm; uniform linear array; wireless system;
  • fLanguage
    English
  • Journal_Title
    Microwaves, Antennas & Propagation, IET
  • Publisher
    iet
  • ISSN
    1751-8725
  • Type

    jour

  • DOI
    10.1049/iet-map.2008.0188
  • Filename
    5161686