【作者】 范崇祎；

【导师】 黄晓涛；

【作者基本信息】 国防科学技术大学 ， 信息与通信工程， 2012， 博士

【摘要】 工作在VHF/UHF频段的低频合成孔径雷达（SyntheticAperture Radar，SAR）能够穿透叶簇进行探测成像，是隐蔽目标侦察与监视的重要手段。具备运动目标指示（Ground Moving Target Indication，GMTI）能力的低频SAR/GMTI系统能够探测开阔地带裸露和树林隐蔽目标、静止与运动目标，具备大范围侦察和监视能力，已成为当前SAR领域内的研究热点之一。与传统高频SAR/GMTI系统相比，低频SAR/GMTI系统信杂比低，多使用体积较大、成本较高的阵列天线。目前，对体积小、成本低的单/双通道低频SAR/GMTI技术研究相对较少。为推进低频SAR/GMTI系统的实用进程，本文结合工程实际，研究了单/双通道低频SAR/GMTI系统的若干理论和技术问题。本文的主要研究工作可概括如下：1.研究了低频SAR的运动目标及杂波特性，所得结论为本文后续研究提供了理论基础和假设依据。首先，基于SAR运动目标二维频谱，对各类运动目标成像算法进行了统一推导。基于推导结果，分析了方位子视下的运动目标特性，为提高实际系统中运动目标信杂比奠定了基础。其次，基于低频SAR/GMTI实测数据，研究了适合低频系统的杂波统计分布模型，完善了干涉图的G分布模型族，验证了树干随方位视角变化的电磁散射特性及叶簇杂波内部运动的相关结论。2.研究了基于大积累角的运动目标检测算法。针对现有运动目标检测算法存在计算量大或运动目标关联困难的问题，充分利用低频系统方位积累角大的特点，提出了基于多方位分辨率图像的运动目标检测算法、基于方位子视图像差的运动目标检测算法和低虚警的运动目标快速检测流程。所提算法具备较高的杂波抑制能力，提高了运动目标的信杂比，且算法计算量小，适用于机载低频SAR/GMTI系统。3.研究了基于部分多普勒信息的运动目标定位算法。首先，针对慢速运动目标，提出了适用于低信杂比的精确定位算法，解决了“方位位置不确定”问题，为实际系统完成运动目标在SAR图像上的高精度标定奠定了基础。其次，针对快速运动目标，提出了解多普勒模糊与抗干扰尖峰两种多普勒中心估计方法，提高了低频SAR/GMTI系统对空中目标的监视性能。4.研究了正交频分复用（Orthogonal Frequency Division Multiplexing，OFDM）信号在低频SAR/GMTI系统中的应用。首先，推导了OFDM SAR的模糊函数，从理论上证明了信号编码对成像分辨率影响不大。其次，研究了编码项在实际成像过程中的影响，提出了基于窗函数加权与距离向编码项补偿的方法，抑制了距离向旁瓣，提高了OFDM SAR方位压缩性能。仿真表明，所提方法能够使OFDM信号达到与传统线性调频信号相当的成像性能，解决了OFDM SAR成像的理论问题。最后，为了克服低频系统的低信杂比和树干引起的多径效应，提出了基于OFDMSAR的杂波抑制及运动目标检测模型，实现了单通道SAR在距离-多普勒域的杂波抑制。该方法计算量小、实时性高，适用于对地面目标进行快速检测及跟踪，检测性能优于线性调频信号。5.研究了适用于机载低频SAR/GMTI系统的通道预处理技术。首先，针对双通道低频SAR/GMTI系统干涉图特有的射频干扰（Radio Frequency Interference，RFI）抑制问题，提出了基于距离频率-方位时间域干涉图像的RFI抑制方法。该方法能消除实测数据中弱RFI对干涉图的影响，避免通道间独立操作引入的额外误差，弥补了现有RFI抑制方法的不足。其次，根据运动误差补偿技术、RFI抑制方法和低频系统通道误差特点，提出了适用于机载双通道低频SAR/GMTI系统干涉图像的通道预处理方法。该方法能显著改善双通道沿航向干涉图的相干性，利于沿航向干涉（Along Track Interferometric，ATI）取得良好的检测效果。本文研究成果已成功应用于某机载低频SAR/GMTI系统实测飞行试验数据处理，对推进低频SAR/GMTI系统实际应用与发展具有重要的参考价值。本文取得的研究成果不仅可应用于多通道低频SAR/GMTI系统，也可为高频SAR/GMTI系统研制提供参考。更多还原

【Abstract】 The low frequency Synthetic Aperture Radar (SAR) working at VHF/UHF bandcan imaging targets under foliages, which is an important way for the detection andsurveillance of conceal targets. The low frequency SAR system with the capability ofGround Moving Target Indication (GMTI) can detect both static and moving targets inthe open or under foliages, which has become one of hot points of SAR research field.Comparing with conventional high frequency SAR/GMTI system, low frequencySAR/GMTI system has lower Signal to Clutter Ratio (SCR) and is common to use thearray antenna with larger size and higher cost. At present, there is little research on thesingle or dual-channel low frequency SAR/GMTI system, which has smaller size andlower cost. To promote the practical application of low frequency system, sometheoretical and technical problems of single or dual-channel low frequency SAR/GMTIsystem are studied with the real data in this dissertation. The main contents of thisdissertation are summarized as follows:1. The characteristics of moving target and clutter in low frequency SAR arestudied. The conclusions in this part provide both a theoretical basis and a basis ofassumptions for the following research in this dissertation. Firstly, based on the derivedtwo-dimensional spectrum of moving target in SAR image, the unified derivation ofmoving target imaging algorithms is deduced and the characteristic of moving target atdifferent azimuth sub-look images is studied. The research results show the potential toimprove SCR of moving target in practical system. Secondly, based on thelow-frequency SAR/GMTI measured data, the clutter statistical distribution model inlow-frequency system is studied, the interference pattern of the G distribution modelfamily is perfected, and the electromagnetic characteristics varying with azimuthsub-look of the trunk and the internal clutter motion of foliage clutter are verified.2. The moving target detection algorithms based on the wide accumulation angleare proposed. Heavy computation load and the associated difficulties of moving targetin a sequence of sub-look images are the problems of the practical processing with theexisting low-frequency moving target detection algorithms. Taking full advantage of thewide accumulated angle in low-frequency system, two moving target detectionalgorithms based on multi-resolution images and the difference of sub-look images areproposed respectively. A novel moving target fast detection flow with low false alarm isproposed in later. These proposed detection algorithms have the good capability ofclutter suppression and can enhance the SCR of moving target, which are suitable forthe practical application in the airborne low frequency SAR/GMTI system.3. The moving target positioning methods based on the part of Doppler informationare proposed. Firstly, a precise positioning method is proposed to locate slow moving targets in low SCR. It can solve the problem of azimuth position uncertainty and lay thefoundation for the practical system to achieve the high-precision calibration of themoving target in SAR images. Secondly, two velocity estimation methods are proposedto solve the ambiguity of Doppler spectrum and anti-peak interference for fast movingtargets. They can improve the potential performance of surveillance of air targets inlow-frequency SAR/GMTI system.4. The Orthogonal Frequency Division Multiplexing (OFDM) signals in thesingle-channel low-frequency SAR/GMTI system is studied. Firstly, the ambiguityfunction of OFDM SAR is derived, which shows that the signal encoding has littleinfluence on imaging resolution. Secondly, the impact of the OFDM encoding term inthe actual SAR imaging processing is studied, and two algorithms based on a weightedwindow and the encoding compensation are proposed respectively, to suppress thesidelobe in range and to improve the azimuth compression performance of OFDM SAR.The proposed latter algorithm can achieve the similar imaging performance with that ofthe traditional linear frequency modulation signal, which solves the theoretical problemsof OFDM SAR imaging. Thirdly, a model of clutter suppression and moving targetdetection is built to overcome the low SCR and the multipath effect from the trunk inlow frequency system. This model achieves high performance of clutter suppression inthe range Doppler domain by single channel SAR. The detection method has a lowcomputation load and is suitable for real-time processing, which can be used in rapiddetection and tracking of ground moving targets. The performance of detection is betterthan that of the linear frequency modulation signal.5. The pre-processing techniques for the airborne low frequency SAR/GMTIsystemare are studied. Firstly, a Radio Frequency Interference (RFI) suppressionmethod in the range-frequency azimuth-time domain of interferogram is proposed forthe RFI of along track interferogram in the dual-channel low-frequency SAR/GMTIsystem. This suppression method can eliminate the weak RFI in the interferogram fromthe real data, avoid the extra error from independent operation between channels andmake up for the drawback of existing RFI suppression methods. Secondly, combiningwith the motion error compensation techniques, the proposed RFI suppression methodand the characteristics of errors between low-frequency channels, a completepre-processing flow is proposed for the airborne dual-channel low-frequencySAR/GMTI system. This flow can greatly improve the coherence of along trackinterferogram in dual-channel system and help Along Track Interferometry (ATI)achieve good detection results.The satisfied research results in this dissertation have been testified by the flyingexperiment data from an airbone low frequency SAR/GMTI system, They haveimportant references value to advance the practical application and development oflow-frequency SAR/GMTI system. The proposed methods can be used not only in multiple channel low frequency SAR/GMTI system, but also as the references to thedesign of high frequency SAR/GMTI system.更多还原