基于修正熵的距离扩展目标检测器

Range-spread target detector based on modified entropy

针对采用球不变随机向量(spherically invariant random vector,SIRV)建模的非高斯杂波背景下,目标散射点个数估计失配时,修正的依赖于散射点密度的检测器(modified scatter density dependent,MSDD)检测距离扩展目标存在一定的信噪比损失问题,提出了一种基于修正熵的MSDD检测器(modified entropy-MSDD,ME-MSDD)。ME-MSDD先采用修正熵对目标强散射点个数进行估计,然后对目标强散射点进行积累形成检测统计量。最后,针对不同的目标模型及实测目标一维距离像对检测器性能进行评估,结果表明,与不依赖于散射点密度的检测器(non-scatter density dependent,NSDD)、基于动态阈值的检测器(dynamic threshold,DT)和MS-DD相比,ME-MSDD能根据实际观测值自适应地估计目标散射点个数,具有更好的检测性能和鲁棒性。

The modified scatterer density dependent （MSDD） has a certain loss of the signal-to-clutter ratio （SCR） when falsely estimating the number of scatterers in non-Gaussian clutter modelling by spherically invariant random vector （SIRV）. For this problem, a modified scatterers density dependent detector based on modified entropy （ME-MSDD） is addressed. For the ME-MSDD, it estimates the number of strong scatterers by using modified entropy firstly, and then the test statistics are formed by integrating strong scatterers. Finally, the detection performance is evaluated for different target models and the one dimension range profile of real targets. The results show that the ME-MSDD can adaptively estimate the number of target scatterers according to the observations, which can improve the performance and robustness of detectors effectively.