IC-PSO算法的收敛性分析及应用研究  

Convergence Analysis of IC-PSO Algorithm and Its Application Research

作　　者：朱奇光[1] 王洪瑞[2] 

机构地区：[1]燕山大学信息科学与工程学院,河北秦皇岛066004 [2]河北大学电子信息工程学院,河北保定071002

出　　处：《光电工程》2010年第4期108-112,共5页Opto-Electronic Engineering

基　　金：国家自然科学基金资助项目(60877047);科技部国际科技合作项目(2008DFR10530)

摘　　要：针对标准PSO算法后期迭代搜索效率不高,容易陷入局部最优的问题,提出将免疫克隆(IC)原理引入PSO算法中,把抗体视为粒子,根据亲和度的高低进行粒子克隆选择、克隆抑制和高频变异,提高了种群的多样性和全局搜索的能力。并将其应用于40Gb/s的传输系统中进行了DOP优化补偿实验,算法补偿所需时间约为71ms。通过对比补偿前后的信号眼图可以发现,PMD补偿后,信号眼图张开度有明显改善,证明了算法的有效性。Considering that the standard （Particle Swarm Optimization） PSO algorithm has low iteration efficiency during later period and may trap to local optimum, Immune Clone （IC） principle is introduced into the PSO algorithm. The antibodies can be regarded as the particles. According to the degree of affinity, the clone selection, clone suppression, and high-frequency mutation are performed, which can enhance the diversity of particle swarm and the capability of global searching. The optimal compensation experiment is performed in the 40 Gb/s transmission system, in which the compensation time required was about 71 ms. The opening of signal eye diagram has been improved obviously after compensation. The experimental results demonstrate the effectiveness of the algorithm proposed.

关 键 词：群智能 免疫克隆 粒子群优化 偏振模色散补偿 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]