基于聚类算法的地面目标HRRP分割识别  

HRRP Recognition for Ground Targets Based on Clustering Algorithm

作　　者：范颖锐 罗小波[1] 宋志勇[1] 付强[1] 

机构地区：[1]国防科学技术大学电子科学与工程学院ATR重点实验室,长沙410073

出　　处：《全球定位系统》2015年第6期58-63,82,共7页Gnss World of China

基　　金：国家自然科学基金(批准号:61401475)

摘　　要：地面目标往往随机排列密集分布,雷达波束内不可避免地存在多个目标,为采用一维距离像进行地面目标识别带来了困扰。本文将波束内多目标距离像的相互关系划分为四种情况,在采用一维距离像进行分割提取的基础上,引入了方位角度信息,仿真结果表明:提出的二维聚类算法有效实现了波束内多目标距离像的分离,同时,本文研究了因这种分割算法造成的距离像残缺率对地面目标识别性能的影响。The ground targets are often randomly arranged in dense distribution,inevitably more than one of them appear in the radar beam,so it brings trouble to the ground target recognition for HRRP.In this paper,the relationship between the multi-target HRRP in the rader beam is divided into four cases,the author introduces the azimuth angle information to the HRRP segmentation,The simulation result shows that the proposed 2Dclustering algorithm can effectively achieve the separation of the multi-target HRRP in the beam,meanwhile,this paper researches the effect of HRRP loss ratio caused by segmentation algorithm for ground target recognition.

关 键 词：一维距离像 二维聚类 残缺率 地面目标 

分 类 号：TN957[电子电信—信号与信息处理]