色噪声环境下宽带非高斯信号的测向算法  

A method of direction finding for non-Gaussian wideband signals in the presence of unknown colored noise fields

作　　者：刁鸣[1] 陈娟[1] 袁熹 

机构地区：[1]哈尔滨工程大学信息与通信工程学院,黑龙江哈尔滨150001

出　　处：《哈尔滨工程大学学报》2009年第9期1066-1069,共4页Journal of Harbin Engineering University

摘　　要：针对色噪声环境下宽带非高斯信号的DOA估计问题,提出了一种去噪与高阶累积量相结合的DOA估计新方法.该方法首先估计出噪声分量,尽可能去除噪声;然后对去噪信号进行聚焦,构造出聚焦后信号的四阶累计量矩阵.对各累计量矩阵的平均值做特征分解,再进行一维谱峰搜索,即可得到来波方向的估计值.该方法不仅适用于高斯色噪声,也适用于非高斯色噪声环境,且不需要知道噪声的原始分布信息.实验结果证明了该算法的有效性.To provide direction of arrival （DOA） estimation for non-Gaussian wideband signals in the presence of unknown colored noise fields, a new method was proposed which combines de-noising and a higher-order cumulant. The first step was to estimate the noise component and remove it. Next, the output array vectors at different frequency bins were transformed into the focusing frequency by a focusing operation, and this determined the individual cumulant matrix. Then the DOA was estimated using the subspace method after eigen-decomposition of the average cumulant matrix. This method is not only applicable to Gaussian colored noise, but also to non-Gaussian colored noise. Furthermore, information about the original distribution of the noise is not needed. Extensive simulations demonstrated that the algorithm has good performance.

关 键 词：色噪声 四阶累积量 非高斯信号 宽带测向 频域聚焦 

分 类 号：TN911.7[电子电信—通信与信息系统]