用于单比特稀疏偶极子阵列波达角估计的协方差矩阵估计方法  

Covariance matrix estimation method for DOA estimation of one-bit sparse cross-dipole array

作　　者：程智勇 陈胜垚[1] 吴文[1] 刘中[1] Cheng Zhiyong;Chen Shengyao;Wu Wen;Liu Zhong(School of Electronic and Optical Engineering,Nanjing University of Science andTechnology,Nanjing 210094,China)

机构地区：[1]南京理工大学电子工程与光电技术学院,江苏南京210094

出　　处：《南京理工大学学报》2022年第3期321-330,共10页Journal of Nanjing University of Science and Technology

基　　金：国家自然科学基金(62171224)。

摘　　要：为了提高单比特稀疏偶极子阵列的波达角(DOA)估计精度,提出了一种协方差矩阵估计算法。利用偶极子阵列结构特征,将该阵列的接收信号重排为具有相同阵列排布的等效标量阵接收信号,其理想协方差矩阵具有托普利兹结构。利用该矩阵的托普利兹结构以及单比特采样信号与无量化接收信号的符号一致性,将等效标量阵接收信号协方差矩阵的估计问题转化为具有等式约束的非凸优化问题,并将此问题松弛为凸问题,求得该矩阵的估计值,进而使用子空间方法获得DOA的估计结果。实验结果表明:相较于单比特多重信号分类(OBMUSIC),在嵌套阵上,该文单比特稀疏协方差矩阵估计(OB-SCME)的估计精度最小均方误差(MSE)有平均2.2 dB的性能提升;在互质阵上,其估计精度MSE平均提升了8 dB;当信噪比为0且快拍数少于60时,其估计精度提升可达10 dB以上。In order to improve the estimation accuracy of direction of arrival(DOA)for one-bit sparse cross-dipole arrays,a covariance matrix estimation method is proposed here.Using the structural characteristics of a dipole array,the received signals of the array are rearranged into equivalent scalar array received signals with the same array arrangement,and its ideal covariance matrix has Toplitz structure.Using the Toplitz structure of the matrix and the symbol consistency between the one-bit sampled signal and the unquantized received signal,the estimation problem of the received signal covariance matrix of the equivalent scalar array is transformed into a nonconvex optimization problem with equality constraints.The optimization is also convexly relaxed and is solved by semi-definite programming.The sub-space method is acted on the estimated covariance matrix to solve the DOAs.The experimental results show that compared with one-bit multiple signal classification(OBMUSIC),on the nested matrix,the estimation accuracy minimum mean square error(MSE)of one-bit sparse covariance matrix estimation(OB-SCME)has an average performance improvement of 2.2 dB;on the coprime array,its estimation accuracy MSE is improved by 8 dB on average;when the signal-to-noise ratio is 0 and the number of snapshots is less than 60,the estimation accuracy can be improved by more than 10 dB.

关 键 词：单比特采样 稀疏阵列 偶极子 波达角 协方差矩阵估计 托普利兹结构 单比特多重信号分类 

分 类 号：TN911.7[电子电信—通信与信息系统]