Currently, sparse Bayesian learning (SBL) has been introduced to solve direction of arrival (DOA) estimation in different situations. In the line of DOA estimation under impulsive noise, existing SBL-based methods need large computation which will restrict their practicabilities. To address this problem, we propose an efficient method based on a real-valued SBL for non-circular signals in this paper. Firstly, received signal model is transformed into a real-valued form using the characteristic of non-circular signals' structure. Then, a sparse representation of the modified signal model is constructed in the presence of impulsive noise. Finally, SBL is applied to reconstruct the real-valued sparse model and solve the DOAs estimation. A series of simulations are carried out in different conditions to evaluate the proposed method. Simulation results demonstrate that our method shows better performance than existing methods.

当前，已经引入稀疏贝叶斯学习（SBL）来解决不同情况下的到达方向（DOA）估计。在脉冲噪声下的DOA估计中，现有的基于SBL的方法需要大量计算，这将限制其实用性。为了解决这个问题，本文提出了一种基于实值SBL的非圆形信号有效方法。首先，利用非圆形信号结构的特点，将接收信号模型转化为实值形式。然后，在存在脉冲噪声的情况下构造修改信号模型的稀疏表示。最后，将SBL应用于重建实值稀疏模型并求解DOA估计。在不同条件下进行了一系列仿真，以评估所提出的方法。