星载InSAR轨道误差建模与分析

发布时间：2018-06-30 05:36

本文选题：合成孔径雷达干涉测量 + 基线估计　；参考：《中国地质大学(北京)》2017年硕士论文

【摘要】：星载合成孔径雷达干涉测量(Synthetic Aperture Radar Interferometry,InSAR)是一种重要的大范围形变监测技术。不同误差源会使测量结果产生偏差,当形变作为主要信息时,InSAR数据处理的目的是消除或者减弱其他信号的影响。轨道误差以近似线性趋势存在于地形高程图和形变图中。目前常用的减弱轨道误差的方法是对残余相位做快速傅里叶变换计算残余基线误差;或者,对残余相位做二次曲面拟合。这两种方法对多数情况是适用的,但是不适合大气传播延迟效应严重的情况,也不适合大范围的形变信号如构造运动或者潮汐等和轨道误差有相同表现特征的情况。时序InSAR技术出现之后,采用最小二乘等方法计算干涉图中的基线误差残余相位,利用不同时空基线的组合,根据平差原理,计算准绝对轨道误差,对轨道精度进行交叉验证来评价其可靠性。也有研究建立模型联合解算形变速率和轨道误差,还有研究认为轨道误差在时间上没有相关性,在时序InSAR技术中可以忽略轨道误差的影响。但是,对于高精度的形变监测,轨道误差对结果的影响不可忽略,因而对轨道误差进行建模和分析是十分必要的。本文根据目前基线误差和轨道误差建模与分析的现状,针对InSAR中的轨道误差进行了研究,主要工作如下:1.详细阐述了基于轨道信息、干涉图信息和地面控制点信息的三种基线估计方法的原理、适用条件和实现过程,用Matlab编写程序实现了三种方法。应用高分三号数据对基线估计的三种方法进行了分析验证,分析了不同方法的效果和适用性。2.提出了基于地面控制点相位残差定权的基线估计方法,推导了算法的数学模型,结合模拟数据论证了该方法在一定程度上减弱了大气传播延迟相位和形变相位对基线估计的影响。3.利用轨道误差网平差原理,实现了对Los Angeles地区的Envisat ASAR数据的轨道误差网平差,利用数据探测的方法,剔除观测值中的粗差,获得较为可靠的轨道误差。
[Abstract]:Spaceborne synthetic Aperture Radar Interferometry (InSAR) is an important deformation monitoring technique. Different error sources will cause deviation of measurement results. When deformation is the main information, the purpose of InSAR data processing is to eliminate or weaken the influence of other signals. The orbit error exists in the topographic elevation map and the deformation map as an approximate linear trend. At present, the commonly used method to reduce orbit error is to calculate the residual baseline error by fast Fourier transform on the residual phase, or to do the Quadric surface fitting to the residual phase. These two methods are suitable for most cases, but they are not suitable for serious atmospheric propagation delay effects or for large scale deformation signals such as tectonic motions or tides with the same characteristics as orbital errors. After the time series InSAR technology appeared, the residual phase of baseline error in interferogram was calculated by least square method, and the quasi-absolute orbit error was calculated according to the principle of adjustment by using the combination of different space-time baselines. The track accuracy is cross-validated to evaluate its reliability. There are also some studies on the joint calculation of deformation rate and orbit error, and the conclusion that orbit error has no correlation in time, and the influence of orbit error can be ignored in time series InSAR technology. However, for high precision deformation monitoring, the influence of orbit error on the result can not be ignored, so it is necessary to model and analyze orbit error. According to the current situation of baseline error and orbit error modeling and analysis, the orbit error in InSAR is studied in this paper. The main work is as follows: 1. The principle, applicable conditions and realization process of three baseline estimation methods based on orbit information, interferogram information and ground control point information are described in detail. The three methods are realized by programming with Matlab. Three methods of baseline estimation are analyzed and verified by using the data of high score three. The effects and applicability of different methods are analyzed. A baseline estimation method based on phase residuals weight of ground control points is proposed, and the mathematical model of the algorithm is derived. Combined with the simulation data, it is demonstrated that this method weakens the influence of atmospheric propagation delay phase and deformation phase on baseline estimation to a certain extent. The orbit error net adjustment of Envisat ASAR data in Los Angeles region is realized by using the principle of orbit error net adjustment. By using the method of data detection, the gross error in the observed value is eliminated and the more reliable orbit error is obtained.
【学位授予单位】：中国地质大学(北京)
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：P207.1;P237

【相似文献】