基于平滑l_0范数和TSVD的MIMO雷达目标参数估计方法  

The Method of MIMO Radar Target Parameter Estimation Based on SL0 Algorithm and Truncated Singular Value Decomposition

作　　者：陈金立[1,2] 周运[2] 李家强[1,2] 朱艳萍[2] CHEN Jin-li ZHOU yun LI Jia-qiang ZHU Yan-ping(Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044, China College of Electronic and Information Engineering, Nanjing University of Information Science and Technology, Nanjing 210044, China)

机构地区：[1]南京信息工程大学气象灾害预报预警与评估协同创新中心,南京210044 [2]南京信息工程大学电子与信息工程学院,南京210044

出　　处：《中国电子科学研究院学报》2017年第3期295-301,共7页Journal of China Academy of Electronics and Information Technology

基　　金：国家自然科学基金(No.61302188;61372066);江苏省自然科学基金(No.BK20131005);江苏高校优势学科Ⅱ期建设工程资助项目;江苏省博士后科研资助计划(No.1402167C)

摘　　要：利用平滑l_0范数(Smoothed l_0,SL0)算法估计MIMO雷达目标参数时,在设定初始值和计算梯度投影中需要对呈病态的感知矩阵进行求伪逆运算,然而病态矩阵的伪逆精度较低,从而导致SL0算法无法直接用于估计MIMO雷达的目标参数。为此,本文提出了一种基于SL0算法和截断奇异值分解(Truncated Singular Value Decomposition,TSVD)的MIMO雷达目标参数估计方法。该方法对感知矩阵进行SVD变换,并设定特征值均值为截断门限,保留大于门限的特征值及其对应的左奇异向量和右奇异向量,并利用SVD反变换获得条件数较小的非病态感知矩阵,实现了SL0算法在MIMO雷达目标信号重构问题中的应用。实验结果表明,与迭代加权lq算法相比,本文方法在保证目标信号重构性能的基础上,明显提高了MIMO雷达的目标参数估计速度。For the estimation of MIMO radar target parameters with the smoothed l0 （ SL0 ） norm algo- rithm, the pseudo inverse operation of ill-conditioned sensing matrix is required in setting initial value and calculating gradient projection. Due to the low pseudo inverse accuracy of ill-conditioned matrix, the SLO algorithm cannot be directly used for estimating target parameters in MIMO radar. Thus, a method based on SL0 algorithm and truncated singular value decomposition for MIMO radar target parameter esti- mation is proposed in this paper. The SVD transform is performed on the sensing matrix, and then the mean value of the singular values is selected as a truncation threshold. The singular values over the threshold are retained, as well as the corresponding left singular vectors and fight ones. Finally, the SVD inverse transform is utilized to obtain a non ill-conditioned sensing matrix with small condition number. Thus the application of SL0 algorithm for MIMO radar target signal reconstruction is achieved. Experi-mental results demonstrate that the reconstruction performance of the proposed method that of the iterative reweighted lq minimization method, while significantly improving the is almost same to speed of parame-ters estimation in MIMO radar.

关 键 词：MIMO雷达 SL0算法 病态矩阵 截断奇异值分解 

分 类 号：TN957[电子电信—信号与信息处理]