论文部分内容阅读
提出利用文中构造的循环互相关矩阵进行特征空间分解，并根据特征值的相对大小进行源数估计的方法．该方法适用于宽带和窄带源，并可较好地解决相干源的源数估计问题．用该方法进行源数估计在波达方向（ＤＯＡ）估计中基本不增加运算量，而且在低信噪比下，仍能准确地进行源数估计．用蒙特卡洛试验证实了该方法的源数估计性能优于通常的源数估计方法．
This paper proposes a method to decompose the feature space by using the cyclic cross-correlation matrix constructed in the paper and estimate the source number according to the relative size of the eigenvalues. The proposed method is suitable for wideband and narrowband sources and can solve the problem of estimating the number of sources of coherent sources. The estimation of the number of sources by this method does not increase the computational complexity basically in DOA estimation, and the estimation of the number of sources can still be done accurately with low SNR. Monte Carlo experiments show that this method has better performance than the usual source estimation method.