基于遗传算法的压缩感知DOA测量矩阵设计.pdfVIP

西安邮电大学学报 Vo l. 21 No.6 第 21 卷第6 期 JOURNAL OF XIAN UNIVERSITY OF POSTS AND TELECOMMUNICATIONS Nov. 2016 2016 年 11 月 doi: 10. 13682/j. issn. 2095-6533. 2016. 06.018 基于遗传算法的压缩感知 DOA 测量矩阵设计 杨洁，刘朋露 (西安邮电大学通信与信息工程学院，陕西西安 710121) 摘 要:基于格拉姆矩阵(Gram matrix) ，利用遗传算法给出一种优化压缩感知波达方向测量矩阵的方法。对阵元 进行不重复采样，构造测量矩阵E 根据稀疏字典得到压缩感知 DOA 的感知矩阵，并由其构造格拉姆矩阵z 以所得格 拉姆矩阵非对角元素的平方和作为遗传算法的适应度值，利用遗传算法求解出最佳采样阵元。仿真实验显示，以 所给方法设计的测量矩阵，在采样阵元较少且信噪较低的情形下，仍具有较好的 DOA 估计性能。 关键词:遗传算法;压缩感知;波达方向s 测量矩阵F 格拉姆矩阵 申图分类号: TN91 1. 7 文献标t只码:A 文章编号: 2095-6533(2016)06-0093-05 Design of compressive sensing DOA measurement matrix based on genetic algorithm YANG Jie , LIU Penglu (佳hool of Communiætion and Information EI粤neer吨， Xì an University of Rωts and Telecommuniætions ，觅an 710121 ， α1Ïna) Abstract: Based on Gram matrix , a method of optimizing the measurement matrix of compressive sensing DOA is presented by using genetic algorithm. Against array elements to carry out non repeated sampling , and construct the measurement matrix. According to the sparse dictionary , the sensing matrix of compressive sensing DOA is obtained , and the Gram matrix is constructed therewith. Take the sum of squares of the non-diagonal elements in the constructed Gram matrix as the fitness value of the genetic algorithm , and use this algorithm to pick out the Optimum sampling array element. Experimental simulation results show that , even in case of less sampling array element and low signal noise ratio , the measurement matrix given by the presented algorithm is of good DOA estimation performance. Keywords: gen