A Binary Channel Characterization Using Genetic Algorithm

スポンサーリンク

概要

• 論文の詳細を見る

• This paper presents an estimation method based on genetic algorithm (GA) for use in estimating the Gilbert's model parameters from an experimental error-gap distribution. A ranking-based selection scheme is introduced to alleviate the problem of premature convergence in GA optimization process. In order to strike a better balance between exploration and exploitation for global convergence, the GA is further improved by applying an SA-based mutation. The results of the conducted experiments indicate that for channel characterization the GA with SA-based mutation yields the near-global optimum solutions that are consistent from run to run.

• 社団法人電子情報通信学会の論文
• 1998-01-20

著者

• Tan T‐h

  Department Of Communication Engineering National Chiao Tung University

• Tan Tan-hsu

  Department Of Communication Engineering National Chiao Tung University

• Chang Wen-Whei

  Department of Communication Engineering National Chiao Tung University

関連論文

• AUTOMATED CRYPTANALYSIS OF BANDSPLITTER-BASED SPEECH SCRAMBLERS
• A Binary Channel Characterization Using Genetic Algorithm
• Memory-Enhanced MMSE Decoding in Vector Quantization(Speech and Hearing)
• Perceptual-Based Playout Mechanisms for Multi-Stream Voice over IP Networks

スポンサーリンク