基于直接数据域方法的快速波达方向估计算法

Fast DOA Estimation Algorithm Based on Direct Data Domain Approach

提出一种基于直接数据域方法的快速波达方向估计算法,该法仅采用阵元输出电压的极少数快拍数据即完成对相干与非相干入射信号的波达方向估计。首先根据直接数据域最小二乘思想构造矩阵方程,采用共轭梯度法求得各观察方向的输出功率,然后得到方向谱。最后利用输出功率粗估和方向谱细估,从而确定入射信号的波达方向。它无需构造协方差矩阵及进行矩阵求逆,又避免了复杂度较高的广义特征值分解。仿真结果证实了算法的估计性能以及相对传统算法所体现出的优越性。

A fast direction of arrival （DOA） estimation algorithm is presented based on direct data domain least square （D^3LS） approach. The algorithm is used to estimate the DOAs of incident coherent and incoherent signals with very few snapshot data of the voltages measured at array elements. Firstly, the matrix equation is formed according to the concept of D^3LS. Secondly, the equation is solved by using the conjugate gradient method to obtain the output power from each look direction and the directional spectrum. Finally, with a rough estimation by output power and a fine estimation by directional spectrum, the DOAs of incident signals are determined. The algorithm processes very few snapshot data opposed to forming sample covariance matrix and computing matrix inversion. Moreover, it avoids the generalized eigenvalue decomposition. Simulation results show its estimation capability and advantages over traditional algorithms.