基于模型稀疏表示的单快拍波达方向鲁棒估计

Robust single-snapshot DOA estimation based on sparse representation of array output model

基于传感器阵列输出模型的稀疏重构,研究了利用单快拍数据进行波达方向(Direction of Arrival,DOA)估计的问题。考虑到在实际应用中,目标信号个数远小于传感器阵元数,目标信号DOA相对于空间来说也是稀疏的,将传统的传感器阵列输出模型进行稀疏化表示,得到阵列输出数据的稀疏表示模型,研究了一种基于l1-范数最小化的单快拍DOA估计算法(L1-Min)。该算法将稀疏参数求解问题转化为二阶锥规划(Second-Order Cone Programming,SOCP)问题的一般形式,并在二阶锥规划的框架下求解,同时分析了算法中正则化参数的选取依据。L1-Min算法对小样本、相干多径信号、目标信号角度间隔小等非理想条件都具有较好的鲁棒性。仿真实验验证了算法的有效性。

The issue of single-snapshot direction of arrival(DOA)estimation is researched based on the sparse representation of array output model.In practical application,the number of targets is far less than the number of array elements,and the DOAs of signals are sparse in the space,so the conventional array output model can be reconstructed as a sparse representation model.The the single-snapshot DOA estimation algorithm based on the l1-norm minimization(L1-min)is proposed.The algorithm translates the optimization problem of sparse parameters into a second-order cone programming(SOCP)framework.The selection criterion of the regularization parameters in this approach is analyzed.The proposed algorithm shows an improved robustness to limited snapshots,coherent sources,and closely spaced sources.Simulations show the effectiveness of the L1-min algorithm.