基于DPCA的SAR地面运动目标检测技术研究

发布时间：2018-03-29 20:38

本文选题：合成孔径雷达　切入点：地面动目标检测　出处：《国防科学技术大学》2014年硕士论文

【摘要】：随着雷达成像技术的不断发展,合成孔径雷达地面动目标检测(SAR-GMTI)已经发展为合成孔径雷达应用的一个重要分支。多通道SAR-GMTI技术由于增加了通道数量,可以获得更多的信息来源,已经继单通道SAR-GMTI后成为GMTI研究的主流和热点方向,并在军事情报监视系统中取得了实用性突破。目前已提出的多通道SAR-GMTI技术主要包括偏置天线相位中心技术(DPCA)、沿航迹干涉合成孔径雷达技术(ATI-SAR)、空时自适应信号处理(STAP)技术以及近几年提出的多输入多输出(MIMO-SAR)技术,其中又以前两项技术的研究最为广泛,后两者虽然理论可行却工程实现困难。到目前为止DPCA和ATI-SAR依然是多通道SAR-GMTI领域最有效可行的解决方案,对其改进、完善的研究工作也从未中断。DPCA和ATI-SAR采用近似相同的数据采集系统,其区别主要在于后端信息处理方式不同,对DPCA和ATI-SAR的研究主要集中在复图像域。复图像域DPCA技术用杂波对消后的幅度信息进行目标检测,ATI-SAR用干涉相位信息进行目标检测。虽然其技术方案很清晰,但是这两种方法涉及的一些理论及技术仍然是不完善的。例如,DPCA和ATI-SAR虽然被同时研究多年,但是它们检测性能的受限因素和相互之间的优缺点至今是模糊的,没有一个统一的认识。另外,ATI-SAR的杂波干涉相位和干涉幅度分布研究早已较为成熟,而杂波DPCA幅度分布还鲜有研究,这就限制了自适应检测在DPCA检测中的应用。而且由于受到通道失衡、平台偏航等因素的影响,DPCA和ATI-SAR的检测性能并不十分理想,有待进一步改进。基于以上分析,本文首先改进了双通道复数据建模模型,基于此模型详细分析了影响DPCA和ATI-SAR检测性能的限制因素,并对两种方法的检测性能进行了全面的分析对比。其次,从多视和地物均匀性两个方面分析推导了复图像域DPCA杂波统计模型,给出了参数估计方法,使得CFAR检测在DPCA动目标检测中的应用成为可能。最后,基于有目标区域和无目标区域的区域相关性不同,提出了一种“区域相关比”加权DPCA检测方法,该方法对通道失衡和干涉相位偏移不敏感,仿真实验和实测数据验证均表明该方法检测性能优于传统DPCA方法。
[Abstract]:With the development of radar imaging technology, SAR ground moving target detection (SAR-GMTI) has become an important branch of synthetic Aperture Radar (SAR) application. Multi-channel SAR-GMTI technology can obtain more information sources because of increasing the number of channels. After single-channel SAR-GMTI, it has become the mainstream and hotspot of GMTI research. The multi-channel SAR-GMTI technology proposed at present mainly includes the phase center technology of biased antenna, the ATI-SARG technique along the track interference synthetic aperture radar (SAR), and the space-time adaptive signal processing. And the multi-input and multi-output MIMO-SAR technology proposed in recent years, Among them, the first two technologies are the most widely studied, although the latter two are feasible in theory, but the engineering implementation is difficult. Up to now, DPCA and ATI-SAR are still the most effective and feasible solutions in the field of multi-channel SAR-GMTI. The perfect research work has never stopped. DPCA and ATI-SAR adopt approximately the same data acquisition system, the difference mainly lies in the difference of the back-end information processing method. The research on DPCA and ATI-SAR is mainly focused on complex image domain. DPCA technology in complex image domain uses the amplitude information of clutter cancellation to detect target. ATI-SAR uses interferometric phase information to detect target. However, some of the theories and techniques involved in these two methods are still imperfect. For example, although both DPCA and ATI-SAR have been studied simultaneously for many years, their detection performance constraints and the advantages and disadvantages of each other are still vague. In addition, the research on the phase and amplitude distribution of clutter interference in ATI-SAR has long been mature, but the amplitude distribution of clutter DPCA is seldom studied. This limits the application of adaptive detection in DPCA detection. Moreover, due to the imbalance of channel, platform yaw and other factors, the detection performance of DPCA and ATI-SAR is not very ideal, which needs further improvement. In this paper, we first improve the two-channel complex data modeling model. Based on this model, we analyze the limiting factors that affect the detection performance of DPCA and ATI-SAR in detail, and analyze and compare the detection performance of the two methods. The statistical model of DPCA clutter in complex image domain is deduced from the aspects of multi-view and ground object uniformity, and the method of parameter estimation is given, which makes the application of CFAR detection in DPCA moving target detection possible. Based on the difference of the region correlation between the target region and the non-target region, a weighted DPCA detection method of "region correlation ratio" is proposed, which is insensitive to channel imbalance and interference phase shift. Simulation experiments and experimental data verify that the detection performance of this method is better than that of the traditional DPCA method.
【学位授予单位】：国防科学技术大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN957.52

【相似文献】