基于协方差矩阵重构的互质阵列DOA估计方法

DOA Estimation Based on Coprime Array of Covariance Matrix Reconstruction

针对现有基于互质阵列的DOA估计方法,由于舍弃差联合阵列中非均匀虚拟阵元而导致最大可估计信号损失的问题,提出了基于协方差矩阵重构的互质阵列DOA估计方法。根据差联合阵列与波程差一一对应的关系,将协方差矩阵扩展为一个部分元素缺失的Toeplitz矩阵;通过求解迹范数最小化问题实行对协方差矩阵中零元素的填充,阵列自由度得到提升;利用求根MUSIC方法实现DOA估计。仿真结果表明:该方法有效提高了可估计信号数且能够有效避免传统稀疏重构算法中模型失配对估计性能的影响。

In the existing DOA estimation methods based on coprime array,discarding the non-uniform virtual elements in the difference coarray leads to the decrease of the number of the resolvable signals.In order to solve the problem,a DOA estimation based on covariance matrix reconstruction was proposed.Firstly,according to the characteristics of the difference coarray was corresponding to the wavepath difference,the covariance matrix was expanded toToeplitz matrix,in which some elements were zero.Then,the zero elements were filled by solving the trace minimization problem,thereby the available degree of freedom were increased.Finally,the reconstructed covariance matrix with root music was employed to estimate DOAs.Numerical experiment results demonstrated the proposed method could detect more signals,and alleviated the effect of basis mismatch.