多特征子空间波形优化设计方法  

A Method of Waveform Design Based on Multi Eigen-Subspace

作　　者：纠博[1] 刘宏伟[1] 何学辉[1] 吴顺君[1] 

机构地区：[1]西安电子科技大学雷达信号处理国家重点实验室,西安710071

出　　处：《电子与信息学报》2009年第12期2858-2863,共6页Journal of Electronics & Information Technology

基　　金：教育部长江学者和创新团队支持计划(IRT0645);国家自然科学基金(60772140);国家部委基金资助课题

摘　　要：针对宽带雷达方位不确定的目标识别波形优化设计问题,该文在色噪声背景下提出一种多特征子空间方法,简称为MES方法。它通过各类目标在各个方位下的回波的可分性进行分析,选择多个可以较好体现各类目标回波差异的特征向量,张成多个特征子空间,然后将期望信号投影到这些子空间上形成优化波形。仿真表明,相对于已有方法,该优化方法能更加平衡地增大每一个方位下目标的可分性,从而有效地提高了目标的识别率。Considering issue of target-aspect sensitivities in the waveform design for the recognition of broadband radar targets, a novel method termed Multi Eigen-Subspace （MES） is proposed in the additional of colored noise. The optimization is done via selecting a lot of eigenvectors which can represent the difference between the echoes of different classes of target in all aspect adequately and spanning them to several eigen-subspaces, then the optimized waveform will be obtained by mapping the desired waveform to those eigen-subspaces. The experimental results prove the efficiency of the proposed method. Compared to the available approaches, the MES waveform helps to increase the class separability and obtain the better performance.

关 键 词：目标识别 波形优化 特征子空间 互信息 

分 类 号：TN958[电子电信—信号与信息处理]