基于联合子带ANM的小快拍宽带DOA估计方法

In the detective and estimation passive system, the traditional direction-of-arrival (DOA) estimation methods cannot work directly with deficient snapshots. To solve this problem, an improved wide band DOA estimation approach is developed by applying the atomic norm minimization (ANM) technique to the joint sub-band data. First, every sub-band data of the whole bandwidth focuses on the reference frequency band to obtain the new sub-band data. Next, the new data matrix is reconstructed by the joint sub-band data after rotation. Then, an optimal Toeplitz matrix is obtained from ANM in the form of semi-definite programming. The accurate signal subspace can be acquired by performing eigenvalue decomposition of the retrieval Toeplitz matrix, which avails to DOA estimation. Simulation results show that the proposed wide band DOA estimation method exhibits better performance in the case of deficient snapshots and can estimate coherent source targets.