基于虚拟基线变换和RBFNN的宽频段DOA估计

提出利用虚拟基线变换的方法解决了频率高端的相位模糊，克服由于相位模糊性造成的方向特征的多值模糊问题，从而简化了从方向特征到波达方向的映射关系。提取无模糊的方向特征，训练RBF神经，进行宽频段波达方向估计。理论分析和试验表明，该解模糊方法计算量小，能有效消除圆阵列的相位模糊，适用阵元数大于等于5的任意均匀圆阵列天线（6元除外）；利用解模糊后的特征建立的DOA估计模型结构简单，性能大大优于直接建模和相关干涉仪法，具有优越的性能和工程实用性。

Broadband Direction of Arrival Estimation Based on Virtual Baseline Transformation and RBFNN

The technique of virtual baseline transformation is proposed to solve the phase ambiguity in the paper.It overcomes the multi-valued problem in the high frequency band effectively,and the mapping from directional feature to DOA becomes very simple.Then the unambiguous directional feature is used to train the Radial Basis Function Neural Network(RBFNN) for estimating the direction of arrival in broadband.The analysis and experimental results show that this method has a little computing burden and can avoid the phase ambiguity effectively.Further,the method is proved to be applicable to any uniform circular array(UCA) whose elements are equal or more than 5(expect 6);The structure of DOA estimation model training with ambiguous directional features is simple,and the performance of model is better than modelling directly and correlation interferometer.So the proposed DOA estimation approach has a broad application future.