利用免疫克隆粒子群混合算法实现自适应PMD补偿  

Realization of adaptive PMD compensation using hybrid IC-PSO algorithm

作　　者：朱奇光[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》2007年第26期178-180,248,共4页Computer Engineering and Applications

基　　金：国家自然科学基金(the National Natural Science Foundation of Chinaunder GrantNo60377002)

摘　　要：提出了一种免疫克隆算法与粒子群优化(PSO)算法相结合的混和算法来进行函数优化,克服了PSO算法容易陷入局部极值的不足,通过免疫克隆算法的应用,提高了种群的多样性,增强了算法全局搜索的能力,提高了收敛速度和精度。仿真结果表明,该算法在搜索多维函数最优解中具有优良的性能。将该混和算法应用到光纤偏振模色散(PMD)补偿系统中,补偿后,眼图恢复效果很好,补偿系统的搜索时间最短可达71ms,响应恢复小于20ms,实现了对光纤通信系统的实时自适应PMD补偿。A hybrid algorithm combining Immunity Clone（IC） algorithm with Particle Swarm Optimization（P SO） algorithm is proposed and applied to optimize functions,which overcomes the shortcoming of converging to local optimum for PSO algorithrn.By introducing the IC algorithrn,the diversity of particle swarms and the converging rate and accuracy is increased,and the capability of global searching is enhanced.Frorn the simulation results, it is shown that the hybrid algorithm has perfect property in rnul- ti-dimension function searching.The algorithm can be adopted in adaptive polarization mode dispersion compensation systern,and after compensation,the recoverage effect of eye diagram is obviously good.The searching time of the compensation system can be attained to 71 ms and the response time is less than 20 rns,which can achieve the real time adaptive PMD compensation.

关 键 词：自适应补偿 粒子群优化(PSO)算法 免疫克隆(IC)算法 偏振模色散(PMD) 偏振度(DOP) 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]