【作者】 严浩；

【导师】 刘梅；

【作者基本信息】 哈尔滨工业大学 ， 信息与通信工程， 2010， 硕士

【摘要】 二十世纪九十年代美国提出了一种新的天基SAR体制——分布式SAR或星座SAR,这种雷达体制立即成为SAR领域研究新热点,受到世界各个国家的重视。这种雷达体制在理论上可以克服当前合成孔径雷达不可逾越的困难:无法同时获得宽测绘带和高分辨率SAR图像;SAR-GMTI过程中不能获得足够长的天线孔径以满足最小可检测速度的要求;SAR干涉测高不能有足够长的空间基线以满足人们需要的测高精度或则不能得到相关性足够好的干涉数据。而分布式SAR可以在单行过的情况下同时获得宽测绘带和高分辨率图像,以及得到长的多空间基线进而可以得到高的测高精度并满足最小可检测速度的要求。但是分布式SAR面临很多实现上的困难,例如空间阵列超稀疏分布、回波信号存在距离或/和多普勒模糊、存在阵列位置(基线)测量误差、发射功率限制等等。目前尚无在运行的分布式SAR系统,所以必须克服这些困难才能把这种雷达体制由理论变为工程实践。本文研究了分布式SAR动目标检测系统的实现问题。主要内容可以归纳如下:(1)介绍了SAR的基本原理和成像算法,包括合成孔径原理和当前常用的雷达成像算法(RD算法、CS算法、NCS算法)。分别利用RD算法、CS算法进行点目标成像仿真和机载SAR雷达真实数据成像,然后仿真了NCS算法点目标成像。研究了SAR动目标的的回波特性,分析了动目标对SAR回波多普勒中心频率和多普勒调频率的影响以及对成像结果的影响。(2)研究了如何仿真分布式SAR回波数据。首先研究了SAR面目标回波模拟的问题,从时域法和频域法分别对SAR面目标回波数据模拟进行了仿真。因为时域法是描述了SAR系统接收回波过程的方法,因此利用时域法进行了分布式SAR回波模拟。同时提出了一种基于空间非同时采样原理的分布式SAR回波模拟方法,并将动目标回波数据叠加到这种方法模拟的回波数据中,得到了含有动目标的分布式SAR回波数据。(3)研究了多通道多像素联合动目标检测方法。因为基于空间非同时采样原理仿真的分布式SAR回波数据具有较高的相关性,因此利用该数据进行动目标检测处理。多通道多像素联合动目标检测是一种在图像域检测动目标的STAP算法。利用多个通道的多个像素构造杂波协方差矩阵,并利用最优空时自适应原理构造滤波权值,最后进行最优滤波并得到动目标检测结果。更多还原

【Abstract】 A new space-borne SAR system—Distributed SAR or Constellation SAR was put forward by the United States in the 1990s. This SAR system immediately became a new research hotspot of the radar domain, and was emphasized all around the world. In theory, this radar system could overcome the insurmountable difficulties of the traditional SAR system: The traditional SAR cannot acquire wide swath and high resolution SAR image simultaneously; The traditional SAR-GMTI system cannot acquire long enough baseline to meet the requirement of minimum detectable velocity; The traditional InSAR system cannot acquire long enough space baseline to meet the requirement of height measurement accuracy or cannot obtain high related interferometry data. Distributed SAR could acquire wide swath and high resolution SAR simultaneously and acquire higher height measurement accuracy and lower minimum detectable velocity at single pass. But there are many difficulties to solve to realize the Distributed SAR system, such as highly sparse array distribution, Range and/or Doppler blur of echo signal, baseline measurement error of distributed array, transmitting power limit. There is not Distributed SAR system at work at present, so we have to overcome these problems to translate the theoretical system to engineering realization.In this master thesis, Ground Moving Target Indication is studied for Distributed SAR system. The main results of this thesis are listed as follows.(1) This dissertation introduced the basic theory and imaging algorithm, consisting of the principle of synthetic aperture and SAR imaging algorithm(RD algorithm, CS algorithm, NCS algorithm); Simulated a point target SAR raw data at side looking mode and deal it with RD algorithm and CS algorithm; Simulated a point target SAR raw data at squint mode and deal it with NCS algorithm. And deal the real airborne SAR data with RD algorithm and CS algorithm. The echo character of SAR moving target was researched, consisting of the influence of SAR moving target to the Doppler center frequency and Doppler modulation frequency of SAR echo and to SAR image.(2) This paper researched the method of simulating the raw echo data of Distributed SAR. Firstly, the method of simulating SAR real scene raw signal was researched. SAR real scene raw data was simulated with the time domain method and the frequency domain method. Then used the time domain method simulating the Distributed SAR real scene raw data for the dealing process of the time domain method being analogous to the real SAR process of transmitting and receiving signal. The dissertation put forward a method of simulating Distributed SAR based on the non-simultaneous space-time samples, and synthesized this data and the moving point target raw data to get the Distributed SAR real data containing moving target.(3) The dissertation researched the multi-channel and multi-pixel adaptive signal processing in image domain. Here we used the simulated raw data based on the non-simultaneous space-time samples to simulate the process of Ground Moving Target Indication for the raw data of different SAR being high related. The multi-channel and multi-pixel adaptive signal processing use the multi-pixel of multi-channel construct the clutter covariance matrix and use the principle of optimal space time adaptive processing and spatial steering vector to construct filtering weight and get the result of GMTI.更多还原