基于IPSO-ICA算法的盲多用户检测  

Blind multiuser detection based on IPSO-ICA algorithm

作　　者：刘广威 葛海波[1] 程浩[1] 张杰[1] 

机构地区：[1]西安邮电大学电子工程学院,陕西西安710061

出　　处：《电视技术》2016年第2期23-26,共4页Video Engineering

基　　金：陕西省教育厅自然科学研究基金项目(2011JM8038)

摘　　要：为了有效抑制CDMA通信系统中的多址干扰,提出了一种将新改进的粒子群(IPSO)算法应用到基于独立分量分析(ICA)的盲多用户检测中的方法。首先在IPSO算法中,使用自适应非线性递减惯性权重w降低早熟概率,利用拉伸技术打破早熟现象,从而使最优解更加精确。然后将IPSO算法融合到ICA算法中,得到IPSO-ICA算法来进行盲多用户检测。最后仿真结果表明,在相同通信环境下,该盲多用户检测比已有的基于Fast ICA和PSO-ICA算法的盲多用户检测的误码率更小,收敛速度更快。In order to effectively suppress multiple access interference in CDMA system, a new method that improved particle swarm optimization （IPSO） algorithm is applied to blind multiuser detection which is based on independent component analysis （ICA）. Firstly, in IPS0 algorithm, using nonlinear regressive w reduces the probability of premature, and using stretching techniques breaks premature phenomenon to get more accurate optimal solutions. Secondly, IPSO is used to ICA algorithm to obtain IPS0-ICA algorithm for blind multiuser detection. Finally, the simulation results show that the blind multiuser detection has a smaller bit error rate and faster convergence speed in comparison with the blind muhiuser detection which is based on FastICA and PSO-ICA algorithm under the same communication environment.

关 键 词：盲多用户检测 粒子群优化 独立分量分析 拉伸技术 

分 类 号：TN914.4[电子电信—通信与信息系统]