基于四阶累积量扩展孔径的宽带信源测向算法

Direction finding algorithm for wideband sources with the aperture extension based on fourth-order cumulant

当阵列过载时,常用高分辨测向算法将会失效.针对此问题,提出了一种新的宽带信源测向算法.该算法利用基于最小冗余直线阵列的四阶累积量来扩展阵列孔径.将得到的四阶累积量去除冗余信息后,构造了一新的Toeplitz矩阵.新矩阵充分利用了原协方差矩阵的信息,避免了冗余信息.通过理论分析和计算机仿真,对扩展后阵列孔径、测向性能、检测概率及可分辨信源数进行了对比研究.结果表明,相比相同阵元数的最小冗余直线阵和均匀直线阵,该算法具有大得多的阵列孔径和更高的分辨率,利用M阵元,最多可以分辨M2-M个信源.

When the array is overloaded, the commonly used high-resolution direction finding algorithm becomes invalid. Aiming at this problem, a novel direction finding al- gorithm for wideband sources was proposed. The algorithm used fourth order cumulant based on minimum-redundancy-linear-array （cum-MRLA） to expand the array aperture. After removing the redundant information in the obtained fourth-order cumulant, it con- structed a new Toeplitz matrix. The new matrix made full use of the information inclu- ding in the original covariance matrix, and avoided the redundant information. Through theoretical analysis and computer simulation, the extended array aperture, the direction finding performance, detective probability and the most identifiable source number were studied. Results show that, compared with the minimum-redundancy-linear-array （MR- LA） and uniform linear array with the same element number, the proposed algorithm has much larger array aperture and higher resolution, usingM elements, which canidentify M2 -M sources.