基于一维散射中心匹配的雷达目标识别

本文提出了一种新的基于目标一维散射中心匹配的雷达目标识别方法.该方法在计算两目标匹配函数之前,先根据目标尺寸大小在目标中心附近设定一个距离窗,将位于此距离窗以外的散射中心剔除,以减轻目标区域以外虚假散射中心的影响.对剩下的目标散射中心,根据两目标散射中心之间的距离,对目标之间的散射中心配对,然后定义两目标的匹配函数为所有配对散射中心的"匹配能量"和与两目标所有散射中心能量和的比值.对五类目标缩比模型的外场测量数据进行分类识别实验,结果表明该方法具有良好的目标识别能力,而且对加性高斯白噪声和雷达带宽不敏感.

Radar Target Classification Based on One Dimensional Scattering Centers Matching

This paper presents a new approach to radar target classification based on one-dimensional scattering centers matching.This approach firstly excludes the spurious scattering centers outside of target region using the range window around target region.Then the scattering centers of two targets are associated in pairs according to their relative ranges.The matching function of two targets is defined by the ratio of the total "matching energy" of all the matched scattering center pairs and the total energy of the two targets' scattering centers.The performance of the proposed approach is examined through classification experiments.The results show that this approach has good target classification capability,and it is robust to additive Gauss white noise and radar bandwidth.