【作者】 何志华；

【导师】 梁甸农；

【作者基本信息】 国防科学技术大学 ， 信息与通信工程， 2011， 博士

【摘要】 分布式卫星合成孔径雷达(SAR)是将卫星编队和星载SAR技术有机结合的新体制天基雷达系统,该系统能够获得比多SAR简单组网更多的功能和更高的性能,是全天候、全天时、高效率获取全球高精度无缝隙地面三维数字高程信息(DEM)的优选手段,成为目前国内外的研究热点。但由于该系统关键技术多、系统集成测试复杂、各误差源间存在紧密的耦合关系,仅通过全数字仿真验证已不能完全满足分布式卫星SAR系统集成验证要求,而采用分布式卫星SAR半实物仿真验证技术是解决这一难题的有效途径。本文瞄准分布式卫星SAR半实物仿真这一前沿迫切课题,围绕“信号协同误差建模”、“半实物回波模拟”和“半实物试验应用”这三个半实物仿真关键技术问题展开研究。各章具体内容安排如下:第二章研究了分布式卫星SAR信号协同层面相位和时间同步方法,以及信号同步误差的误差模型及其对InSAR性能的影响。分析了相位同步误差对InSAR性能的影响;建立了直达波和双向脉冲交换这两种基于同步信号传递的相位同步方法的补偿相位信号模型,对比分析了各类剩余相位误差的特性和量级;研究了乒乓双站模式下的相位同步处理方法,提出一种在回波域利用相关处理直接进行相位同步处理的方法;研究了时间同步误差对干涉测绘带损失、双站SAR成像和InSAR测高的影响;研究了基于GPS驯服高稳晶振的时间同步方法;建立了通道频域幅相模型及其干涉信号模型,研究了通道一致性误差对干涉相位偏差和标准差的影响。第三章研究了分布式卫星SAR数字回波信号高精度高效仿真方法。提出一种分布式卫星SAR数字回波信号高精度快速仿真算法,将回波信号表示成发射信号与场景调制信号的卷积,采用升采样技术确保回波信号仿真精度,通过将场景调制信号降采样到雷达工作采样频率保证了半实物回波模拟过程可在模拟器上实时实现,该算法具有模型适用性广、回波仿真精度高、回波计算效率高的优点;提出回波仿真计算量概念,可用于定量评估回波仿真算法性能和硬件计算能力;研究了基于时空分解的并行回波高性能计算方法和基于GPU加速的回波高性能计算方法。第四章针对分布式卫星SAR半实物仿真验证需求,完成了双通道回波信号模拟器的设计与实现。完成了回波模拟器的总体设计,完成了射频子系统、中频子系统、数字处理部分和显控软件的设计与实现,最终完成回波模拟器研制;提出一种基于相关加窗的模拟器高精度幅相特性估计方法,能够有效抑制闭环数据中噪声和杂散的影响;研究了复系数FIR滤波器设计及优化方法,可根据估计出的幅相特性设计出复FIR滤波器系数;完成了复系数FIR滤波器的FPGA实时实现。经测试回波模拟器的工作模式和通道幅相特性达到对回波模拟器的指标要求。第五章研究了分布式卫星SAR相位、时间同步与通道一致性误差的半实物仿真与评估。给出半实物仿真系统中数字仿真系统的体系结构和实物仿真系统的组成;提出了一种基于数据分析的半实物误差特性提取方法;提出了一种基于不同设备连接方案的半实物误差隔离和溯源试验评估方法,设计了分布式卫星SAR相位同步误差半实物验证试验,测试得到相位同步误差对干涉测高的影响,验证了双向脉冲交换相位同步方法的工程可行性;设计了分布式卫星SAR时间同步误差半实物验证试验,测试得到时间同步误差对干涉测高的影响,验证了基于GPS驯服高稳频率源的时间同步方法的工程可行性;设计了分布式卫星SAR通道一致性误差半实物验证试验,测试得到通道一致性误差对干涉测高的影响。更多还原

【Abstract】 Distributed spaceborne synthetic aperture radar (SAR) is an innovative spaceborne radar instrument based on a combination of the satellite formation technique and the SAR technique. It has more function and higher performance than simple integrated SAR constellations and can effectively generate high-resolution global digital elevation model (DEM) in all weather conditions both by day and by night, which has become the hot subject in all over the world. As the system has a lot of key techniques and complicated system integration and testing problem and close coupling among system error sources, the computer-based simulation verification can not completely meet the integration and testing requirements of the distributed spaceborne SAR. Using the hardware-in-the-loop simulation techniques is an effective means to deal with this problem. This thesis focuses on the hardware-in-the-loop simulation of the distributed spaceborne SAR system. The systematic research is carried out about three key techniques of the hardware-in-the-loop simulation in the signal synchronization error modeling, the hardware-in-the-loop raw signal simulation and the hardware-in-the-loop experiment applications. The research in each chapter is arranged as following:The synchronization method, error model and its influence on InSAR performance of the signal synchronization error of the distributed spaceborne SAR is studied in chapter 2. The influence of the phase synchronization error on InSAR performance is analyzed. The compensated phase model of the two phase synchronization methods by synchronous signal transmission, the direct-path echo method and the pulsed alternate method, is presented. And the characteristics and amount of the residual phase synchronization errors are analyzed comparatively. The phase synchronization processing method in alternating bistatic mode is studied, and an echo-domain phase synchronization processing method using correlation processing is proposed. The influence of the time synchronization error on the interferometric swath loss, the bistatic SAR imaging, and the interferometric height measurement is analyzed. The time synchronization method based on GPS disciplined USO (Ultra Stable Oscillator) is studied. A channel amplitude and phase mismatch model in frequency domain and its interferometric signal mode is presented, and the influence of channel mismatch error on the InSAR phase deviation and variance is analyzed.The accurate and effective digital raw signal simulation method of the distributed spaceborne SAR is studied in chapter 3. An accurate and fast raw signal simulation algorithm is proposed. The raw signal is expressed as the convolution of the transmitting signal and the scene modulation signal, and a interpolation technique is used to maintain the accuracy of the fast approximation algorithm. The scene modulation signal is decimated to the radar sampling frequency for real-time implementation in the simulator. The method has advantages such as wide applicability, high simulation accuracy and high computation efficiency. The raw signal calculated quantity concept is proposed to quantitatively evaluate the fast raw signal simulation algorithm performance and the hardware computing capability. A parallel raw signal simulation HPC (High Performance Computing) method based on the time and space decomposition, and a GPU accelerated raw signal simulation HPC method is studied.The design and implementation of the dual-channel raw signal simulator for the hardware-in-the-loop simulation application of the distributed spaceborne SAR is studied in chapter 4. The raw signal simulator is successfully designed and implemented, including the general design, the radio frequency subsystem, the medium frequency subsystem, the digital signal processing section, and the display and control software. A correlation and windowing method is proposed to accurately extract the simulator amplitude and phase frequency response, which can restrain the influence of the noise and spur in the recorded loop data. A complex FIR filter design and optimization scheme is proposed to design the complex FIR filter coefficients according to the estimated frequency response. A FPGA technique is used to implement the filter in real-time. The measured results of the work mode and the channel amplitude and phase frequency response show that the indexes of simulator satisfy the design requirements.The hardware-in-the-loop experiment and evaluation of the phase and time synchronization and the channel mismatch error of the distributed spaceborne SAR is studied in chapter 5. The computer simulation system architecture and the physical simulation system configuration is presented. An error characteristics extraction method based on data analysis is proposed, and an error isolated evaluation method by different hardware configuration is proposed. The hardware-in-the-loop experiment of phase synchronization error is designed to obtain the influence of the error on the InSAR height measurement accuracy, which validates the feasibility of the pulsed alternate phase synchronization method. The hardware-in-the-loop experiment of time synchronization error is designed to obtain the influence of the error on the InSAR height measurement accuracy, which validates the feasibility of the time synchronization method based on GPS disciplined USO. The hardware-in-the-loop experiment of channel mismatch is designed to obtain the influence of the error on the InSAR height measurement accuracy.更多还原