改进粒子群优化在压缩感知DOA估计中的应用  

Improved PSO and Its Application to CS DOA Estimation

作　　者：赵宏伟[1] 刘波[1] 刘恒[1] 

机构地区：[1]西安空间无线电技术研究所,陕西西安710100

出　　处：《微电子学与计算机》2016年第5期33-36,41,共5页Microelectronics & Computer

基　　金：国家自然科学基金项目(61201089)

摘　　要：以压缩感知(CS)的波达方向(DOA)估计为研究背景,针对正交匹配追踪算法用于高维信号稀疏恢复时效率低下问题,提出一种结合粒子群优化(PSO)和正交匹配追踪(OMP)的新DOA估计算法.该算法充分利用了PSO算法解决寻优问题的优势,并结合OMP算法的处理流程,具有运算时间较少、估计误差较低和适用范围广等优势.为了提高算法性能,对标准PSO算法的粒子更新机制和惯性权值等进行了改进.仿真结果证明了方法的有效性.The present direction of arrival （DOA） estimation methods based on the compressive sensing （CS） theory is background at present. Shortcoming of greedy pursuit methods is efficiency to solve sparse recovery. A new algorithm based on Particle Swarm Optimization （PSO） algorithm and orthogonal matching pursuit （OMP） algorithm is proposed. It has less computation time, low estimation error and wide application range, which makes full use of the advantages of PSO and OMP algorithm. To improve algorithm performance, the particle updating system of standard PSO algorithm are improved. The results of simulation experiments show that the proposed algorithm is effective.

关 键 词：波达方向估计 粒子群优化 压缩感知 正交匹配追踪 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]