原子范数空时动目标参数分离估计算法  

Atomic Norm Minimization for Maneuvering Target Parameter Separated Estimation in STAP

作　　者：章涛[1] 郑海芳 张威 来燃 ZHANG Tao;ZHENG Haifang;ZHANG Wei;LAI Ran(Tianjin Key Laboratory for Advanced Signal Processing,Civil Aviation University of China,Tianjin 300300,China)

机构地区：[1]中国民航大学天津市智能信号与图像处理重点实验室,天津300300

出　　处：《现代雷达》2025年第3期48-57,共10页Modern Radar

基　　金：中央高校基本科研业务费中国民航大学资助专项(3122019048)。

摘　　要：文中提出了一种基于参数分离的原子范数空时自适应处理动目标参数估计算法。该方法首先对构建的联合稀疏恢复数据通过双线性变换将加速度和速度参数分离,然后利用低秩矩阵恢复理论将两个参数分别在各自连续的原子集中进行稀疏恢复,并实现了利用半正定规划和交替方向乘子法两种求解方式得到对应的参数子空间,最后通过矩阵束方法求解对应的参数估计值,规避了现有稀疏恢复方法中由于字典失配造成的估计性能降低的问题。实验证明,在信噪比较高的情况下,文中方法估计性能优于已有的基于固定离散化字典的稀疏恢复空时动目标参数估计方法。In this paper,a parameter estimation algorithm for moving targets based on atomic norm space-time adaptive processing is proposed.First,the acceleration and velocity parameters are separated by applying bilinear transformation to the constructed joint sparse recovery data.Next,the low-rank matrix recovery theory is used to sparsely recover the two parameters in their respective continuous atom sets,and the corresponding parameter subspace is obtained by using positive semidefinite programming and the alternating direction multiplier method.Finally,the corresponding parameter estimates are solved by the matrix bundle method.The proposed algorithm avoids the problem of estimation performance degradation caused by dictionary mismatch in the existing sparse recovery methods.Experiments show that in the case of high signal-to-noise ratio,the estimation performance is better than that of the existing sparse recovery space-time moving target parameter estimation methods based on fixed discretization dictionaries.

关 键 词：空时自适应处理 参数估计 字典失配 原子范数 

分 类 号：TN959.73[电子电信—信号与信息处理]