| 摘要: |
| 多输入多输出（MIMO）雷达中部分失效阵元会使得阵列采样数据丢失，从而导致较差的角度估计性能。为此，提出一种基于不完整矩阵因子重构的MIMO雷达角度估计方法。首先，根据协方差矩阵可分解的性质，提取维度较低的矩阵因子，并将协方差矩阵中缺失数据恢复问题转化为矩阵因子重构问题。然后，为了利用矩阵因子中元素的相关性，对不完整矩阵因子建立核范数约束下的低秩Hankel矩阵重构模型；为避免传统的核范数最小化求解中计算复杂度高的问题，采用低秩矩阵拟合方法将Hankel矩阵分解为两个维度较低的矩阵，等价表达了核范数约束。最后，利用交替方向乘子法（ADMM）对该矩阵重构模型进行求解。仿真结果表明，本文方法可以有效地重构出矩阵因子中的缺失元素，进而实现阵列协方差矩阵中丢失数据的补全，改善阵元失效下的MIMO雷达角度估计性能。 |
| 关键词: MIMO雷达 阵元失效 角度估计 Hankel矩阵 矩阵因子重构 |
| DOI：DOI:10.3969/j.issn.1672-2337.2023.06.009 |
| 分类号:TN911.23；TN958 |
| 基金项目:国家自然科学基金（No.62071238）；江苏省自然科学基金（No.BK20191399） |
|
| Angle Estimation Method in MIMO Radar Based on Matrix Factor Reconstruction |
|
CHEN Jinli， JIANG Zhijun， ZHU Xicheng， LI Jiaqiang
|
|
School of Electronics and Information Engineering, Nanjing University of Information Science and Technology, Nanjing 210044, China
|
| Abstract: |
| Partial failed elements in multiple input multiple output（MIMO） radar may result in the loss of array sampling data and thus poor angle estimation performance. To mitigate this performance degradation，a novel MIMO radar angle estimation method is proposed based on incomplete matrix factor reconstruction. The method firstly utilizes the decomposability property of the covariance matrix to extract the low?dimensional matrix factor, transforming the problem of missing data recovery in the covariance matrix into a problem of reconstructing the matrix factor. Afterwards, a low?rank Hankel matrix reconstruction model is established by exploiting the correlation between the elements in the matrix factor and imposing a nuclear norm constraint on the incomplete matrix factor. To avoid the high computational complexity of the traditional nuclear norm minimization problem, this method decomposes the Hankel matrix into two lower?dimensional matrices using the low?rank matrix fitting method. This realizes an equivalent representation of the nuclear norm constraint. Finally, the alternating direction method of multipliers（ADMM） is employed to solve the matrix reconstruction model. Simulation results demonstrate the effectiveness of the proposed method in reconstructing the missing elements within matrix factors, thereby completing lost entries in array sampling data and improving angle estimation performance for MIMO radar under element failure. |
| Key words: MIMO radar array element failure angle estimation Hankel matrix matrix factor reconstruction |
