基于粒子群优化算法的稀疏信号盲分离  

Blind Sparse Source Separation Based on Particle Swarm Optimization

作　　者：高鹰[1] 谢胜利[2] 许若宁[1] 李朝晖[1] 

机构地区：[1]广州大学城广州大学信息与机电工程学院,广州510006 [2]华南理工大学电子与信息学院,广州510641

出　　处：《系统仿真学报》2006年第8期2264-2266,共3页Journal of System Simulation

基　　金：国家自然科学基金(60274006);中国博士后科学基金(2003034062);广东省自然科学基金博士科研启动基金(04300015);广州市科技计划项目(2004J1-C0323);广州市属高校科技计划项目(2055)

摘　　要：把粒子群优化算法应用于稀疏信号盲分离中,采用基于粒子群优化算法的聚类算法来估计混合矩阵;然后利用粒子群优化算法在求解具有线性约束优化问题时,只需在初始化时粒子满足线性约束条件,无须做其它处理的特点来求解稀疏源信号,从而给出了一种基于粒子群优化算法的稀疏信号盲分离算法。该算法计算量小,精度较高。仿真结果表明该算法是有效的,具有良好的分离性能。A blind sparse source separation algorithm was proposed by using the particle swarm optimization. The proposed algorithm first estimates the mixing matrix by using the particle-swarm-optimization-based clustering algorithm, Then the particle swarm optimization which always meets the constraints except for initialsing all particles to be feasible solutions in solving linearly constrained optimization problem is applied to the linearly constrained optimization problem for recovering the sparse sources. The presented algorithm is characterized by high accuracy and less computation. Simulation results illustrate the efficiency and the good performance of the algorithm.

关 键 词：稀疏信号盲分离 粒子群优化算法 聚类 线性约束优化问题 

分 类 号：TN911.72[电子电信—通信与信息系统]