干涉SAR测高测速技术研究

【作者】 穆文争；

【导师】 张晓玲；

【作者基本信息】 电子科技大学 ， 信号与信息处理， 2008， 硕士

【摘要】 干涉合成孔径雷达(InSAR)是合成孔径雷达(SAR)功能的扩展,有两种工作模式:切航迹干涉(CT-InSAR)和沿航迹干涉(AT-InSAR),其中切航迹干涉具有获取地面高程的能力,沿航迹干涉具有检测地面慢速运动目标的能力。由于InSAR测量精度高以及全天候、全天时工作的特点,有着巨大的军用与民用价值,是目前SAR领域研究的热点之一。本文主要研究了InSAR技术中几个关键性的问题,如CT-InSAR中的图像配准、AT-InSAR中的运动目标检测及运动速度估计等,最后研究了混合基线下InSAR的相位分离问题。本文的主要工作有:1、详细阐述InSAR高程测量及运动目标检测的基本原理。给出了CT-InSAR的数学模型,指出在一定近似的条件下,干涉相位和高程之间存在线性关系;然后分析了AT-InSAR运动目标的回波模型,并对已有的检测方法作了分类和简要介绍。2、研究了CT-InSAR自适应图像配准的干涉相位估计方法。该方法的特点是将图像配准和干涉相位估计同时进行。文章在给出其数学模型及处理步骤的基础上,对方法作了改进,并用SIR-C/X-SAR实测数据对改进后的效果进行了验证。3、研究了基于协方差矩阵的运动目标检测方法。分析了小特征值法和非对角元素幅度差法,并提出一种基于协方差矩阵条件数的方法。与传统方法相比,该类方法的优点是杂波抑制效果好,但运算量稍大。4、研究了DPCA和DCFT相结合的运动目标运动速度估计方法。在DPCA检测出运动目标的基础上,用基于能量熵的DCFT估计出其多普勒参数,然后根据这些参数和运动速度之间的关系,可以得到目标的距离向和方位向速度。5、研究了混合基线情况下InSAR的相位分离问题。为了同时实现测高测速,可以将InSAR基线混合偏置,然后将所得到的干涉相位分离即可。文章推导了理论模型并通过计算机仿真对分离的可行性进行了验证,并指出在这种情况下可以先用基于协方差矩阵的运动目标检测方法检测出运动目标,然后再相位分离。更多还原

【Abstract】 Interferometric synthetic aperture radar (InSAR) is function expansion of synthetic aperture radar (SAR). It includes two work modes: cross track interferometry SAR (CT-InSAR) and along track interferometry SAR (AT-InSAR). CT-InSAR can get the goal ground’s elevation, and AT-InSAR can realize ground slow moving target detection. Because of its high precision measurement and may work in all weather and all day, InSAR has great value in military and civil, so it has become one of the focuses of research in the SAR field. This dissertation mainly studies several crucial issues in InSAR technologies, such as image coregistration in CT-InSAR, moving target detection and estimation of motion parameters in AT-InSAR. Finally, Phase separation in InSAR with hybrid baseline is studied.The major work of this dissertation is as follows:1、Basic principles of SAR elevation measurement and ground moving target detection are detailed. First, mathematical model of CT-InSAR is introduced. The dissertation points out that under certain approximative conditions, a linear relationship exists between the interferometric phase and the elevation. Then the dissertation analyzes the model of moving targets’echoes in AT-InSAR, classificates and briefly introduces the detection methods that have been proposed.2、An estimation method for CT-InSAR interferometric phase with adaptive image coregistration is studied. The characteristic of this method is that image coregistration and interferometric phase estimation are done at the same time. The dissertation gives the method’s mathematical model and processing steps, then improves it. Finally, the effectiveness of the improved method is verified with the real data from SIR-C/X–SAR.3、A moving target detection method base on the covariance matrix is studied. The dissertation firstly introduces the methods that have been proposed, such as small values, magnitude difference of non-diagonal elements, then a method based on the covariance matrix eigenvalue ratio is proposed. The dissertation compares these methods with traditional ones, and points out that such method has effective clutter suppression, while computation is larger. 4、An estimation method for moving targets’velocities based on the combination of DPCA and DCFT is studied. The moving targets’Doppler parameters can be estimated with DCFT which is based on energy entropy. Then according to the relationship between these parameters and moving target’s velocities, the range velocity and azimuth velocity can be estimated.5、Phase separation in InSAR with hybrid baseline is studied. In order to obtain elevation and velocities at the same time, we can tilt the baseline, then measure elevation and velocities by separating the hybrid phase. The dissertation derives the theoretical model and verifies the feasibility of this method through a simulation on the computer, then points out that in such case, moving targets can be detected firstly by methods based on the covariance matrix, then the phase can separated.更多还原