采用降秩多级维纳滤波器的二维DOA估计快速算法

摘要

In two-dimensional direction-of-arrival ( DOA) estimation of two parallel linear arrays,a fast method is presented based on multi-stage Wiener filter( MSWF) to reduce the computational complexity. First,signal subspace is obtained quickly by using the forward recursions of the MSWF instead of estimating covariance matrix and eigen decomposition. Then, azimuth and elevation angles are estimated individually by MUSIC algorithm and paired automatically, which converts the 2-D estimation into two 1-D problems and decreases the computational complexity further. The simulation results indicate that the performances of the algorithm are better than that of DOA Matrix Method( DOAM) which is also based on the same array model. The proposed algorithm is also effective when elevation angles are equal and its computational com-plexity is low, so it is suitable for high real-time DOA estimation.%针对双平行线阵的二维波达方向( DOA)估计问题，为有效降低计算复杂度，提出了一种基于降秩多级维纳滤波器( MSWF)的快速算法。首先利用MSWF的前向递推实现信号子空间的快速估计，无需估计协方差矩阵和特征分解；然后，通过MUSIC算法对方位角和俯仰角的估计进行分维估计，使二维DOA估计退化为两个一维DOA估计问题，且方位角和俯仰角自动配对，进一步降低了运算量。仿真结果表明，该方法的估计精度优于同样基于双平行线阵提出的波达方向矩阵法( DOAM)，俯仰角兼并时同样适用，计算复杂度低，适用于实时性要求高的应用背景。