计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (10): 230-233.

• 工程与应用 • 上一篇    下一篇

基于改进型Hopfield神经网络的盲多用户检测算法

王鸿斌1,2,李福平1,张立毅1,王华奎1   

  1. 1.太原理工大学 信息工程学院,太原 030024
    2.忻州师范学院 计算机科学与技术系,山西 忻州 034000
  • 收稿日期:2007-11-13 修回日期:2008-01-24 出版日期:2008-04-01 发布日期:2008-04-01
  • 通讯作者: 王鸿斌

Blind multi-user detection algorithm based on improved Hopfield neural network

WANG Hong-bin1,2,LI Fu-ping1,ZHANG Li-yi1,WANG Hua-kui1   

  1. 1.College of Information Engineering,Taiyuan University of Technology,Taiyuan 030024,China
    2.Department of Computer Science,Xinzhou Teachers University,Xinzhou,Shanxi 034000,China
  • Received:2007-11-13 Revised:2008-01-24 Online:2008-04-01 Published:2008-04-01
  • Contact: WANG Hong-bin

摘要: 基于Hopfield神经网络没有学习规则,不需要训练,也不会自学习,靠Lyapunov函数的设计过程来调节权值的特点,将广义罚函数与Hopfield神经网络的能量函数结合,基于最小平均输出能量准则,构造出更合适的新目标函数,分析讨论了一种实现DS/CDMA盲多用户检测的改进型Hopfield神经网络方法。仿真结果表明,该算法在误码率、抗远近效应方面都有明显的改善。

关键词: 罚函数, 能量函数, 目标函数, 误码率, 远近效应

Abstract: Based on the Hopfield neural network without learning rules,not need training,and not self-learning,to adjust weight by the design process of Lyapunov function,generalized penalty function is combined with the energy function of Hopfield neural network.A more suitable structure of the new objective function is built based on the minimal average output energy norm.An improved Hopfield neural network method of achieving DS/CDMA blind multi-user detection is discussed.Simulation results show that the algorithm is significantly improved in bit error rate and anti-near-far effect.

Key words: penalty function, energy function, object function, bit error rate, Near-Far Effect(NFE)