用于Nested阵列的多测量矢量DOA估计算法

Multi-measurement Vector DOA Estimation Algorithm for Nested Array

Nested阵列协方差矩阵向量化后可认为是单测量矢量(Single Measurement Vector,SMV)模型,基于追踪算法和正交匹配追踪等算法在SMV模型下估计多个信源精度不高。针对这一问题,在空间平滑算法的基础上,提出了一种用于Nested阵列的多测量矢量(Multiple Measurement Vector,MMV)DOA估计算法。将SMV模型转为MMV模型,基于奇异值分解的1范数重构算法进行测向。仿真结果表明,基于所提MMV模型的1范数重构算法能够准确估计出目标数多于阵元个数的信号的波达方向。

The nested array covariance matrix can be regarded as a Single-Measurement Vector(SMV)model after vectorization.The basis pursuit algorithm and the orthogonal matching pursuit algorithm can’t obtain the high accuracy of estimating multiple sources under the above model.To solve this problem,based on the spatial smoothing algorithm,a Multiple Measurement Vector(MMV)Direction of Arrival(DOA)estimation for the nested array is proposed.First,the SMV model is converted to the MMV model,and then the direction finding is performed by the 1 norm reconstruction algorithm based on Singular Value Decomposition(L1SVD).The simulation results show that the 1 norm reconstruction algorithm based on the proposed MMV model can accurately estimate the DOA of the signal when the number of targets is more than that of elements.