基于四阶累积量的测向方法研究

针对MUSIC-LIKE方法能很好地扩展阵列孔径但计算量较大的问题,基于均匀线阵提出了一种快速提取阵列无冗余数据的新方法。该方法基于四阶累积量孔径扩展的原理,摒弃了原MUSIC-LIKE算法在均匀线阵DOA(direction of arrival)估计中的大量冗余数据,从而大大降低了矩阵运算的维数,降低了计算量。经过处理后的四阶累积量具有和二阶统计量类似的形式,因而可以将大多数业已成熟的基于信号二阶统计特性的阵列信号处理方法应用于四阶累积量中,计算机仿真证明了算法的有效性。

Study on DOA estimation based on fourth-order cumulant

A new method which can distill the un-redundant data of uniform linear arrays is proposed,owing to the fact that MUSIC-LIKE algorithm can extend the array aperture but has the large computation cost.The massive redundant data can be eliminated by analyzing the array extension principle of MUSIC-LIKE algorithm to uniform linear arrays.So the dimension of the matrix and the computation cost can be decreased.Having been processed,the form of the fourth-order cumulant matrix is similar to the second-order cumulant matrix,so many mature algorithms based on the signal's second-order statistics can be used in the processing of forth-order cumulant.Its effectiveness is proved by the computer simulation.