卫星导航接收机抗欺骗干扰技术研究

This application aims at researching a challenging and crucial scientific problem, anti-spoofing of navigation satellite receiver. The GPS system of USA and the COMPASS system of China are chosen as study objects. From view of signal processing, the techniques and approaches of automatic detection and recogniton of spoofing are studied, based on the generation principle and the function mechanism of the spoofing signal.The contents of the study include three parts as follows. Firstly, based on the typical hardware structure of a spoofer, the spoofing jamming's model and its function mechanism are studied. Secondly, methods of feature extracion and selection of navigation satellite spoofing are researched. Third, we study the problems of how to detect and recognize the occurence of a spoofing in a receiver. The research is distinguished as: (1) Based on the hardware structure of a spoofer, the subtle effection, by the influence of each component or module, to spoofing signal in the process of generation is analyzed, and then applying mathmatical modelling theory to get behavioral and simulation models, which can accurately describe the dynamic behavior of the jamming. (2) Based on the models, the tiny features of the spoofing signal are extracted and a feature subspace is constructed by feature selection. (3) New detection and recognition methods are proposed, which from statistical signal processing perspective, and the resolution limit is studied. This research will provide theoretical surpport for anti-spoofing design of our country's COMPASS navigation satellite system receivers.

本项目针对卫星导航接收机抗欺骗干扰这一富有挑战性的重要基础科学问题开展研究,以美国GPS/我国北斗卫星导航系统为研究对象，根据卫星导航欺骗干扰的产生与作用机理，从信号处理角度，开展欺骗干扰的自动检测识别新技术与新方法研究。项目研究内容包括：基于典型欺骗干扰机结构的卫星导航欺骗干扰建模与作用机理研究、基于模型的卫星导航欺骗干扰特征提取与特征选择研究、卫星导航欺骗干扰检测识别研究等。创新之处在于：1）从欺骗干扰机硬件结构出发，分析各元器件或模块给干扰信号产生过程造成的细微影响，并基于数学建模理论建立能准确描述欺骗干扰信号动态行为的行为模型和仿真模型；2）基于上述模型，提取欺骗干扰的细微特征并通过特征选择构建约简特征参数子空间；3）从统计信号处理角度提出卫星导航欺骗干扰检测识别新方法并对分辨界开展研究。本项目的研究将为我国北斗卫星导航接收机的抗干扰设计提供理论支持。

本项目针对卫星导航接收机抗欺骗干扰这一富有挑战性的重要基础科学问题开展研究,以美国GPS/我国北斗卫星导航系统为研究对象，根据卫星导航欺骗干扰的产生与作用机理，从信号处理角度，开展欺骗干扰的自动检测识别新技术与新方法研究。项目研究内容包括：基于典型欺骗干扰机结构的卫星导航欺骗干扰建模与作用机理研究、基于模型的卫星导航欺骗干扰特征提取与特征选择研究、卫星导航欺骗干扰检测识别研究等。通过研究，项目组取得的重要结果有：1、在干扰建模方面，研究并比较了5种无记忆和有记忆的欺骗干扰模型，研究结果表明有记忆模型具有更优的建模性能，同时在欺骗干扰识别应用上也具有更优的识别性能。此外，有记忆模型的阶数越高，各方面性都能得到提高，但复杂度也更高。2、在干扰特征提取方面，得到了一些重要且有效的特征参数，如前述有记忆模型的系数向量、Wigner积分双谱奇异值、Wigner双谱对角切片中的多种特征、信道差异特征等；提取特征的方法采用了智能种群算法、经典系统辨识方法、时频分析等并提出了新的方法。3、在欺骗干扰检测识别方面，利用前述特征，研究了基于SVM、神经网络、统计检测和拟合优度的干扰检测识别方法，均取得了良好的识别效果。此外，项目组还在信号分辨界方面开展研究，提出了基于RAO检测的统计分辨界分析方法，具有GLRT检测相类似的渐近性能。上述研究成果，具有较高的理论意义与应用价值，将为我国北斗卫星导航接收机的抗干扰设计提供理论支持和应用支撑。