基于拟合特征值的信源数估计方法  

A Method for Estimating Source Number Based on Modified Eigenvalue

作　　者：蒋敏[1] 黄保涛 伍光新[1] JIANG Min;HUANG Baotao;WU Guangxin(Nanjing Research Institute of Electronics Technology,Nanjing 210039,China)

机构地区：[1]南京电子技术研究所,南京210039

出　　处：《现代雷达》2020年第7期45-49,共5页Modern Radar

摘　　要：针对信源数目估计问题,提出一种基于拟合特征值的信源数估计方法。利用对数几率函数拟合协方差矩阵特征值,改善特征值分布状态,降低信号、噪声特征值集合内离散度。对拟合后的特征值进行差值处理,计算特征值差值的二阶统计量,据此构建信源估计函数估计信源数量。该算法利用特征值的二阶统计量,有效避免了有色噪声背景以及快拍数量对信源数估计的影响。通过数值仿真分析信噪比、快拍数对信源数估计的影响,验证算法的稳定性和有效性。Aiming at the problem of source number estimation, this paper proposes a new method based on fitted eigenvalues. Use the logistic function to fit the eigenvalues of the covariance matrix, improve the distribution of eigenvalues, reduce the dispersion of the signal and noise eigenvalue sets, perform the difference processing on the fitted eigenvalues, and calculate the second-order statistics of difference between the eigenvalues. Based on the second-order statistics a source estimation function is constructed to estimate the number of sources. The algorithm uses the second-order statistics of eigenvalues to effectively avoid the influence of colored noise background and the number of snapshots on the estimation of the number of sources. The effects of the signal-to-noise ratio and the number of snapshots on the estimation of the number of sources are analyzed through numerical simulation to verify the stability and effectiveness of the algorithm.

关 键 词：高分辨 信源数估计 特征值分解 二阶统计量 

分 类 号：TN957.52[电子电信—信号与信息处理]