基于分数阶矩估计的非参量CFAR检测

Nonparametric CFAR Detection Based on Fractional Moment Estimations

该文针对杂波背景中的CFAR(恒虚警率)检测问题,提出了一种新的CFAR检测器——Blind-CFAR检测器。该检测器通过估计杂波的分数阶矩,获得杂波的最大熵概率密度函数估计,进而确定CFAR检测门限。以Weibull杂波为例,仿真验证了Blind-CFAR检测器的有效性。由于采用了非参数估计方法,Blind-CFAR检测器具有结构简单、通用性好等优点,并可应用于未知杂波背景中的目标检测。

For the problem of CFAR（Constant False Alarm Rate） detection in clutter,a new detector,i.e.,the Blind-CFAR detector is proposed.The Blind-CFAR detector calculates its detection threshold through the PDF（Probability Density Function） estimation of clutter,which is based on the fractional moment estimations of clutter and the maximum entropy method.The Blind-CFAR detector is verified in Weibull clutter through simulation experiments.By employing nonparametric estimation methods,the Blind-CFAR detector has advantages of a simple structure,a good universality,etc.Furthermore,it can be applied to target detection in unknown clutter.