复杂环境下雷达信号分选与识别方法研究

In current field of radar countermeasures, there exists following key issues to be solved in sorting and recognizing emitter signals: deeply analyzing substantive characteristics of complex radar emitter signals and exploring the features of sorting and recognizing radar signals in complex environment, and on this basis, studying intelligent model and algorithm for sorting and recognizing radar emitter signals in complex and intensive environment. This project aims at actively studying the effective approaches that resolve existing difficulties in the field of signal sorting and recognition，and attempts to implement explorative and systematic theory research. The researching details include suppressing Gaussian noise and narrowband interference that mixed in intra-pulse broadband information, detection of multi-component radar emitter signals, intrapulse features extraction of radar emitter signals based on the analysis of main ridge characteristics of ambiguity function, feature extraction of full pulse signal based on chaotic oscillator detection, intelligent sorting model and algorithm based on chaotic ant swarm and support vector clustering, novel method for sorting radar signals based on multi-dimensional space biomimetic pattern recognition, performance analysis and evaluation methods of signal sorting model and algorithm. It lays a foundation for establishing novel intelligent sorting and recognition system of radar emitter signals. This research can hopefully provide theory bases for breaking the technical bottleneck of signal sorting and recognition in the field of radar countermeasures.

深入分析复杂体制雷达辐射源信号的本质特征，挖掘复杂环境下的雷达信号分选与识别特征，并在此基础上研究针对复杂密集环境下雷达辐射源信号的智能分选识别模型与算法，是目前雷达对抗领域辐射源信号分选与识别研究工作中亟待解决的关键问题。本项目将积极探讨解决信号分选与识别领域中存在问题的有效途径，并拟进行探索性和系统性的理论研究。具体包括：雷达脉内宽带信息的高斯噪声与窄带干扰抑制问题的研究、多分量雷达信号检测方法的研究、基于模糊函数主脊特征分析的雷达信号脉内特征提取的研究、基于混沌振子检测的全脉冲信号特征提取的研究、基于混沌蚁群联合支持向量聚类的智能分选模型和算法的研究、基于多维空间仿生模式识别的雷达信号分选新方法的研究、信号分选识别模型和算法的性能分析与评价方法的研究。为构建新型智能化的雷达信号分选与识别系统奠定基础。本项目的研究成果有望为突破雷达信号分选与识别的技术瓶颈提供理论依据。

针对日趋复杂、密集的未知雷达信号环境，深入分析和研究了复杂体制雷达辐射源信号的本质特征，探索和挖掘复杂环境下的雷达信号分选特征，并在此基础上研究了针对复杂密集环境下未知雷达辐射源信号的新型分选模型和算法等雷达对抗领域辐射源信号分选研究工作中亟待解决的关键问题。本项目对解决信号分选领域中存在问题的有效途径进行了积极的探索与尝试，并进行了探索性和系统性的理论研究。具体包括：雷达脉内宽带信息的高斯噪声与窄带干扰抑制问题的研究、多分量雷达信号检测方法的研究、基于模糊函数主脊特征分析的雷达信号脉内特征提取的研究、基于混沌振子检测的全脉冲信号特征提取的研究、基于动态数据场聚类的智能分选模型和算法的研究、基于盲分离的MIMO雷达辐射源信号分选算法研究、基于多维空间仿生模式识别的雷达信号分选新方法的研究、基于云模型的信号分选识别模型和算法的性能分析与评价方法的研究。本项目的研究成果将为我国雷达对抗领域的瓶颈技术——信号分选与识别技术提供理论依据，为提高我国电子对抗装备技术水平发挥重要作用。