PSOによるIIRフィルタ設計における局所解停留回避の一検討 (信号処理)  [in Japanese] A study on avoidance of local minimum stagnation in IIR filter design using PSO  [in Japanese]

Search this Article

Author(s)

Abstract

本研究では,PSO (Particle Swarm Optimization)を用いたIIRフィルタ設計について検討する. IIRフィルタの設計問題は非線形最適化問題であり,多数の局所解が存在する.この問題の解法としてPSOが知られている.しかし, PSOは局所解に対する強い指向性ゆえに局所解に停留する場合があり,設計性能が初期値に依存する傾向が強い.本研究では,停留時に目的関数にペナルティを付加し,停留回避する手法について検討する.設計例より,提案法が初期値に依存せずに良好な設計が可能であることを示す.

In this paper, a design method for the IIR (Infinite Impulse Response) filter using the PSO (Particle Swarm Optimization) is developed. The design problem of IIR filters falls into a non-linear optimization problem, and it makes the design a tough job. To solve problem, the method using PSO has been developed. Because the PSO has a strong directivity toward a local minimum, the PSO updating tends to stagnate around such a local minimum and thus indicates a premature convergence. In this study, the method of avoidance of the local minimum stagnation is proposed, in which a penalty is added to an objective function. Several examples are shown to present the effectiveness of the proposed method.

Journal

  • IEICE technical report. Signal processing

    IEICE technical report. Signal processing 113(191), 49-54, 2013-08-29

    The Institute of Electronics, Information and Communication Engineers

Codes

  • NII Article ID (NAID)
    110009805849
  • NII NACSIS-CAT ID (NCID)
    AA11943613
  • Text Lang
    JPN
  • ISSN
    0913-5685
  • NDL Article ID
    024830966
  • NDL Call No.
    Z16-940
  • Data Source
    NDL  NII-ELS 
Page Top