Study On Multibaseline InSAR Phase Estimation Method

Synthetic Aperture Radar Interferometry (InSAR) can realize all-time, all-weather observation the earth, not influenced by light and weather; has huge advantage in fast extracting high-precision, large-scale three-dimensional of the ground, and has already become a hot research in earth observation technology.Currently, the most research is single-baseline InSAR, which can be influenced by baseline, time and terrain easily. In order to overcome the shortcomings of single-baseline InSAR, in1990s, Ferretti, etc, proposed multi-baseline InSAR technology, it can reduce the influence affected by the target height dramatic changes, large noise, terrain un-sampling and layover, achieve higher accuracy and become an important direction in InSAR development. China’s current research on multi-baseline InSAR technology is still shallow, and most of it is based on simulation data. In this paper, we use measured data, such as airborne combination data and satellite track data, to study multi-baseline InSAR phase estimation methods, the research and experiments are explained, and the results of single and multi-baseline InSAR are analyzed and evaluated. The research content and innovation include:The development and applications of InSAR are introduced, and the disadvantages of traditional single baseline InSAR are analyzed. Then, the background, status and key techniques of multi-baseline InSAR are presented, the fundamental of deriving DEM from InSAR are analyzed, and the modes, advantages and processing flowchart of multi-baseline InSAR are discussed.The repeat track InSAR data registration method and matching processes are presented; the geometric form of the baseline is described and the baseline of spaceborne interferometric SAR estimation methods are presented, and then baseline parameters are estimated, which can determine the relative position and ratio between different baselines.This paper studies two classical multi-baseline InSAR phase unwrapping algorithm, they are Least-squares method and Maximum Likelihood method; focuses on the differential filter multi-baseline InSAR phase unwrapping method. In this method, the phase unwrapping of shortest baseline is done. firstly, and then the reference interferogram for longer baseline is calculated using the unwrapped interferogram with shorter baseline and their relationship. Thus, for a long baseline differential interferogram filtering and phase unwrapping will be done.Using airborne combination data and satellite track data, the experiments of phase unwrapping with the three methods for multi-baseline InSAR are finished. Combing the phase unwrapping results, these three methods are compared and analyzed for their applying cases and characters.On the basis of depth study on multi-baseline InSAR phase estimation, this article used VC6.0++to develop a multi-baseline InSAR phase estimation software.