基于最优化方法的地下圆柱目标定位与识别  

Underground Cylindrical Objects Position Estimation and Recognition Based on Optimization Technique

作　　者：雷文太[1] 黄春琳[2] 

机构地区：[1]中南大学信息物理工程学院,湖南长沙410083 [2]国防科技大学电子科学与工程学院,湖南长沙410073

出　　处：《计算机仿真》2009年第7期1-4,共4页Computer Simulation

基　　金：国家自然科学基金(60602022)

摘　　要：地下管道、空洞等圆柱形目标的定位和识别是探地雷达走向实用的一个亟待解决的问题。提出了一种在频域运用最大似然估计进行圆柱目标定位和识别的方法。在一阶Born近似下建立了各孔径点处回波信号和目标的位置参数和电磁参数之间的关系,将目标的未知参数估计问题归结为全局搜索一目标函数极值的最优化问题。通过对仿真数据和实测数据的处理,结果显示该方法可以精确估计出目标的位置参数和电磁参数,有效地进行圆柱目标的定位和识别。In the application of ground penetrating radar, a pivotal problem is the localization and recognition of cylindrical objects such as tracts and cavities. In this paper, a maximum-likelihood estimation method is presented to solve this problem. By using estimation theory, this problem is transferred to the estimation of target＇ s position and electromagnetic parameters. Based on first order Born approximation, the relationship between target＇ s parameters and echo signals is established and a maximum - likelihood estimation method is used to get these parameters. Through processing simulation radargram data and actual radargram data, the algorithm validates it＇ s capabilities in the aspect of GPR application.

关 键 词：探地雷达 目标定位 目标识别 

分 类 号：TN95[电子电信—信号与信息处理]