基于子空间类DOA估计算法具有很高的分辨率,但不适用于相干源测向,提出一种基于多维局部搜索的相干信号源DOA估计算法。首先,对协方差矩阵进行特征分解,获得叠加导向矢量空间。然后,对叠加导向矢量空间进行傅里叶变换,预估计信号方向和延迟系数。最后,以预估计结果为基点,通过多维局部搜索,精确估计相干信号入射方向。由于采用的是局部搜索方式,算法运算量非常小。仿真试验结果表明,提出的算法分辨率高,估计误差小。

Direction-of-arrival (DOA) estimation algorithms using subspace have high resolution. However, they are not applicable to the coherent signals. To settle this problem, a new algorithm using multidimensional local searching is proposed in this paper. First, a superposed steering vector is obtained by performing the eigen-decomposition of the covariance matrix. Then, the DOAs and the delay coefficients are preliminary estimated by the Fourier transform of the superposed steering vector. Finally, the DOAs of coherent signals are exactly estimated by multidimensional local searching based on the results of preliminary estimation. The computational load of the new algorithm is very small for the local searching. The results of simulation experiments show that the new algorithm has high resolution and low estimation error.

国家自然科学基金资助项目