计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (18): 146-148.DOI: 10.3778/j.issn.1002-8331.2010.18.046

• 数据库、信号与信息处理 • 上一篇    下一篇

Lagrange优化的LMK神经网络盲多用户检测

李艳琴1,张立毅2,3,郭纯生1,储琳琳1   

  1. 1.防灾科技学院 防灾仪器系,北京 101601
    2.天津商业大学 信息工程学院,天津 300134
    3.天津大学 电子信息工程学院,天津 300072
  • 收稿日期:2008-12-08 修回日期:2009-02-20 出版日期:2010-06-21 发布日期:2010-06-21
  • 通讯作者: 李艳琴

Lagrange optimized LMK criteria feed-forward neural network blind multi-user detection

LI Yan-qin1,ZHANG Li-yi2,3,GUO Chun-sheng1,CHU Lin-lin1   

  1. 1.College of Disaster Prevention Equipment,Institute of Disaster Prevention Science and Technology,Beijing 101601,China
    2.Department of Information Engineering,Tianjin University of Commerce,Tianjin 300134,China
    3.Department of Electric Information Engineering,Tianjin University,Tianjin 300072,China
  • Received:2008-12-08 Revised:2009-02-20 Online:2010-06-21 Published:2010-06-21
  • Contact: LI Yan-qin

摘要: 提出一种基于LMK准则的前馈神经网络盲多用户检测算法,给出算法的代价函数和约束条件,利用Lagrange方法对带约束的代价函数进行优化,获得前馈神经网络网络权值的迭代公式,从而实现神经网络盲多用户检测。仿真结果表明,新算法具有良好的误码率性能和收敛速度。

关键词: 盲多用户检测, 前馈神经网络, 最小平均峰度(LMK)准则, Lagrange乘子

Abstract: A novel blind multi-user detection algorithm based on Feed-Forward Neural Network(FNN) is proposed.A cost function based on Least Mean Kurtosis criterion and the restraint condition of cost function is founded.Weights and parameters of feed-forward neural network are optimized by Lagrange multiple.Iterative formulas of feed-forward neural network are obtained,and then blind multi-user detection algorithm is realized.Simulation results indicate the novel algorithm possesses lower bit-error ratio,better convergence ability and so on.

Key words: blind multi-user detector, forward neural network, Least Mean Kurtosis(LMK) criteria, Lagrange multiple

中图分类号: