基于粒子群优化算法思想的组合自适应滤波算法  

Combined Adaptive Filtering Algorithm Based on the Idea of Particle Swarm Optimization

作　　者：林川[1] 冯全源[1] 

机构地区：[1]西南交通大学信息科学与技术学院,成都610031

出　　处：《电子与信息学报》2009年第5期1245-1248,共4页Journal of Electronics & Information Technology

摘　　要：根据粒子群优化(PSO)算法的社会心理学指导思想并结合自适应FIR滤波器的特点,设计了合适的惯性项、认知项与社会项表达式,并将之应用于组合自适应滤波器的子自适应滤波器更新中,提出了基于PSO算法思想的组合自适应滤波算法,分析了新算法的计算复杂度。理论分析与不同条件下的自适应系统辨识仿真结果表明,新算法可以在不明显提高计算量的条件下较好地平衡自适应滤波器的稳态失调与跟踪能力,其收敛性能优于其它几种较新的LMS算法。Based on the social psychology idea behind the Particle Swarm Optimization （PSO） algorithm and the feature of adaptive FIR filter, the proper expressions for the ＂inertial＂, ＂cognitive＂ and ＂social＂ parts are designed and applied to the optimization of the adaptive FIR filter in the combined adaptive filter. A combined adaptive filtering algorithm based on the idea of PSO is presented, and the complexity of the new algorithm is also analyzed. The theory analysis azld the simulation results of the adaptive system identification under different conditions show that the new algorithm can balance the steady state misadjustment and tracking ability well, and its convergence performance is better than that of some other new LMS algorithms.

关 键 词：自适应滤波器 粒子群优化 最小均方算法 凸组合 

分 类 号：TN911.7[电子电信—通信与信息系统]