复杂环境下雷达指纹识别系统的关键技术研究

项目名称： 复杂环境下雷达指纹识别系统的关键技术研究

项目编号： No.61203137

项目类型： 青年科学基金项目

立项/批准年度： 2013

项目学科： 自动化学科

项目作者： 王磊

作者单位： 西安电子科技大学

项目金额： 25万元

中文摘要： 随着现代军事对抗信息化特征的日益增强，雷达辐射源指纹（个体）识别已成为当前电子侦察领域的研究热点和难点问题。它通过接收未知雷达辐射源发射的信号，分析其指纹特征，从而唯一地识别辐射源个体，完成准确的搭载平台鉴别和威胁判断。本项目针对实际环境中干扰复杂、信号密度大、雷达参数多变等挑战，开展雷达指纹识别的关键技术研究。主要研究内容有：1.研究雷达辐射源信号预处理方法，包括降噪、多径判别与抑制等；2.研究复杂体制辐射源信号调制分析方法；3.研究辐射源个体特征的产生机理和有效提取方法；4. 研究基于距离准则和模糊核函数优化的特征选择方法，以及基于典型相关分析的特征融合方法；5.研究基于广义置信度评价的信号拒判方法，基于后验概率的分类器融合方法，以及特征库在线更新方法。本项目的研究成果将为进一步完善我国雷达辐射源个体识别的方法体系、促进其工程应用、提高电子对抗系统的性能提供理论与技术支撑。

中文关键词： 雷达辐射源；个体识别；雷达指纹；特征提取；信息融合

英文摘要： With the increasing development of modern military counter system, the research on specific emitter identification (SEI) has become one of the difficulties and hotspots in the field of electronic warfare reconnaissance. By analyzing the intercepted radar signal, it should recognize individual electronic emitters which are of same type or with same parameter. The provided reliable and timely intelligence information could be further used to locate and evaluate the potential radar emitters, as well as the particular units operating them. Aiming at the challenges in real environments such as complex interference, highly dense signals, and changing radar parameters, this project mainly devotes to the core methods for radar SEI. The focuses of this research are as follows: 1. The pre-processing methods for de-noising and identification of multipath signals; 2. Methods for intentional modulation recognition of complex radar signals; 3. Mechanism analysis and extraction of unintentional modulation features; 4. Distance-based and Ambiguity Function guided feature selection, as well as Canonical Correlation Analysis based feature fusion; 5. Signal rejection criteria based on generalized confidence evaluation, classifiers fusion based on the posteriori, and schemes for online updating of radar features. The research findi

英文关键词： radar emitter；specific emitter identification；radarprint；feature extraction；information fusion