改进的智能优化在多用户检测中的应用  

Improved intelligent optimization algorithm used in multiuser detection

作　　者：许成谦[1] 郝红杰[1] 

机构地区：[1]燕山大学信息科学与工程学院,河北秦皇岛066004

出　　处：《计算机工程与应用》2013年第3期143-146,共4页Computer Engineering and Applications

摘　　要：针对概率克隆选择微粒群算法(PCSPSO)在解决离散优化问题时效果不佳的缺点进行改进,将改进后的算法(IPCSPSO)应用于多用户检测,提出基于改进的概率克隆选择微粒群算法的多用户检测器(IPCSPSO-MUD)。IPCSPSO在由二次更新后的记忆集和原种群构成的临时种群中寻找全局最优解,进一步扩大搜索范围;以抗体生存期望值为标准更新种群,保证抗体多样性。仿真结果表明,所提出的多用户检测器在误码率性能、收敛速度、抗远近效应能力和系统容量等方面均有显著提高。To overcome the defect of Probability Clonal Selection Particle Swarm Optimization algorithm（PCSPSO）in solving discrete optimization problems, an improved algorithm（IPCSPSO）is proposed. Then a multiuser detection using IPCSPSO is presented. IPCSPSO searches for globally optimal solution in the temporary population which is made up of twice update’s memory set and original population. Hence the search scope is expanded obviously. To ensure the antibodies diversity, IPCSPSO updates population based on the standard of antibodies survival expectation. Simulation results show that the proposed multiuser detection has significant performance improvement in terms of bit-error-rate, convergence rate, near-far resistance and capacity of system.

关 键 词：码分多址 多用户检测 概率克隆选择微粒群算法 亲和度 抗体生存期望值 

分 类 号：TN914[电子电信—通信与信息系统]