基于噪声雷达的机载前视三维超分辨成像机理与方法研究

The traditional SAR which has been extensively used has commendably realized positive side-looking or front side-looking ground imaging. In recent years ,a common attention has been paid on the study of down-looking 3D ground imaging radar at home and abroad. However, the realization of airborne forward-looking super-resolution 3D ground imaging，for example,the safe landing of the plane and the safe driving of the car in the fog day, remains a difficulty in the research of ground imaging radar studies around the world at present.The noise radar has excellent low intercept probability and anti-jamming features because of the randomness of the transmitted signal.In this project, we will combine the sparse array, the MIMO radar and the noise radar, and conduct the mechanism research of a new kind of airborne or vehicle-mounted forward-looking super-resolution 3D imaging sensor with a good performance of rejecting cross modulation interference, including the optimization design of the sparse array,the forward-looking 3D imaging of static and moving targets,and the super-resolution 3D imaging based on compressed sensing , the theoretical research of forward-looking high-resolution imaging to static and moving targets and so on. The research methods of theoretical analysis, system simulation and simple experimental platform validation will be adopted to solve the problems in basic theories, parameter selections and constructions of the actual system of the forward-looking super-resolution 3D imaging system, and verify theories, algorithms and key technologies of forward-looking 3D imaging. The study of this project could accumulate bases in the aspects of new theories and algorithms for the new airborne or vehicle-mounted forward-looking super-resolution imaging sensor, and breakthrough the corresponding key technologies, which will provide new theoretical and technical supports to improve the capability of vehicle-mounted or airborne forward-looking super-resolution 3D imaging of our country.

传统SAR很好地实现了正侧视或前侧视对地成像，并得到广泛应用。近两年下视对地三维成像雷达研究得到国内外普遍重视，但在雾天或低能见度情形下，如何实现飞机降落和汽车驾驶等需要的前视对地三维超分辨雷达成像是当前国内外研究的难点和工程急需。噪声雷达由于其发射信号的随机性，因而具有十分优良的低截获概率和抗干扰特性。本项目将噪声雷达、稀疏阵列、和MIMO技术相结合，开展一种具有良好抗交调干扰性能的机载或车载前视三维超分辨成像机理研究，其中包括阵列优化设计、前视对静止和运动目标三维成像、超分辨算法推导等。采用理论分析、系统仿真和简易实验平台验证的研究方法，解决相应的基础理论、参数选择和实际系统构建等问题，实现前视三维成像理论、算法和关键技术的验证。本项目研究可在机载或车载前视三维超分辨成像雷达新理论和算法方面积累基础，突破相应关键技术，为提高我国机载或车载前视三维超分辨成像能力提供新理论和技术支撑

传统的机载或星载SAR能够很好地实现正侧视或前侧视对地成像，而忽视了前视成像，因而在许多应用场景中限制了SAR成像技术的发展。稀疏阵列的研究可以在降低系统复杂度的情况下获得满阵相同的空间分辨力。MIMO雷达通过正交发射和匹配滤波器组进行处理来形成虚拟阵元。在抗干扰层面上，噪声雷达的发明使得雷达在低截获表现，抗干扰能力上相比于传统的雷达具有更大的优势。本项目主要将稀疏阵列天线、MIMO技术和噪声雷达技术进行结合来设计雷达成像系统，在弥补传统雷达的缺陷的情况下，实现前视成像。主要研究侧重于稀疏阵列MIMO噪声体制机载或车载前视三维超分辨成像雷达的体制研究、稀疏阵列MIMO雷达正交发射波形优化设计研究和前视成像算法和交轨向超分辨算法研究。在理论方面，研究出了分段式非线性调频的正交频分复用雷达波形，基于Keystone变化，斜视最小化和椭圆波形的前视成像算法，基于改进LBF的MLBF前视成像算法，基于稀疏表示的MIMO逆SAR的成像算法以及一系列关于估计分布源的算法。这些算法均相比于传统算法有较大的提升。在实验方面，设计出了适用于本项目的MIMO稀疏天线阵，MIMO收发单元，MIMO雷达信号综合产生器、综合处理器、信号模拟器等硬件设备。其在室内实验，室外实验等仿真实验中能够获得在距离上大于2000m的探测距离、小于2.5m的探测精度，在角度上获得小于2.58度的测角精度。本项目在前视超分辨成像新理论和算法方面，突破了相应的关键技术，对我国机载或车载前视超分辨成像能力提供了新技术支撑。