中国物理B ›› 2007, Vol. 16 ›› Issue (6): 1619-1623.doi: 10.1088/1009-1963/16/6/022

• GENERAL • 上一篇    下一篇

Discrete channel modelling based on genetic algorithm and simulated annealing for training hidden Markov model

赵知劲, 郑仕链, 徐春云, 孔宪正   

  1. Telecommunication School, Hangzhou Dianzi University, Hangzhou 310018, China
  • 收稿日期:2006-08-27 修回日期:2006-10-27 出版日期:2007-06-20 发布日期:2007-06-20
  • 基金资助:
    Project supported by Pre-Research Foundation of Electronics Science Research Institute (Grant No 41101040102).

Discrete channel modelling based on genetic algorithm and simulated annealing for training hidden Markov model

Zhao Zhi-Jin(赵知劲)†, Zheng Shi-Lian(郑仕链), Xu Chun-Yun(徐春云), and Kong Xian-Zheng(孔宪正)   

  1. Telecommunication School, Hangzhou Dianzi University, Hangzhou 310018, China
  • Received:2006-08-27 Revised:2006-10-27 Online:2007-06-20 Published:2007-06-20
  • Supported by:
    Project supported by Pre-Research Foundation of Electronics Science Research Institute (Grant No 41101040102).

PDF (PC)

646

赞

摘要： Hidden Markov models (HMMs) have been used to model burst error sources of wireless channels. This paper proposes a hybrid method of using genetic algorithm (GA) and simulated annealing (SA) to train HMM for discrete channel modelling. The proposed method is compared with pure GA, and experimental results show that the HMMs trained by the hybrid method can better describe the error sequences due to SA's ability of facilitating hill-climbing at the later stage of the search. The burst error statistics of the HMMs trained by the proposed method and the corresponding error sequences are also presented to validate the proposed method.

Abstract: Hidden Markov models (HMMs) have been used to model burst error sources of wireless channels. This paper proposes a hybrid method of using genetic algorithm (GA) and simulated annealing (SA) to train HMM for discrete channel modelling. The proposed method is compared with pure GA, and experimental results show that the HMMs trained by the hybrid method can better describe the error sequences due to SA's ability of facilitating hill-climbing at the later stage of the search. The burst error statistics of the HMMs trained by the proposed method and the corresponding error sequences are also presented to validate the proposed method.

Key words: hidden Markov model, discrete channel model, genetic algorithm, simulated annealing

中图分类号:  (Numerical optimization)

  • 02.60.Pn
02.50.Ga (Markov processes)