基于CS阵列的DOA估计

基于目标在空域分布稀疏的性质,通过引入压缩感知(Compressive Sensing或Compressive Sampling,CS)理论的思想,提出一种基于奇异值分解的压缩采样阵列(SVD-CSA)DOA估计算法。首先建立DOA压缩感知模型,根据阵列结构建立过完备原子库,然后对压缩采样阵列结构输出的数据矩阵进行奇异值分解,最后基于范数约束的最优化问题的目标函数将信号子空间分解到最佳基向量上,实现了空域信号DOA的高分辨估计。相对于已有算法,该算法减少了硬件复杂度,具有较低的运算量,且能够对相干信号进行有效DOA估计。实验仿真验证了其有效性。

DOA estimation based on compressive sampling array

Based on the sparse property of the targets distributed in spatial domain,the idea of Compressive Sensing（CS） theory is introduced.A compressive sampling array（CSA） DOA algorithm based on the singular value decomposition（SVD） is proposed.Firstly a DOA compressive sensing model is constructed,and an over-complete atom dictionary is established according to the array geometry.Then SVD of the output data matrix of the CSA is completed.Finally a best matched basis vector can be found by decomposition of signal subspace based on the objective function of the optimization issue via norm constraint,it can obtain high-resolution estimation of DOA.Compared with the traditional methods,the proposed method reduces the hardware complexity greatly,has a lower computational complexity,and can also handle the coherent signals.The simulation results verify the effectiveness of the proposed algorithm.