Research On Phase Unwrapping Algorithm Of Timeseries InSAR Based On Network Optimization

Interferometry Synthetic Aperture Radar(InSAR)technology is a new space earth observation technology.Because of its large-scale,high-precision and high-resolution characteristics,it has become one of the important tools for digital elevation model generation and surface displacement monitoring.With the gradual improvement of the time resolution of satellite data,the multitemporal InSAR technology came into being,which not only reduces the error caused by phase decorrelation,but also provides new solutions for other key technologies in InSAR data processing.In InSAR technology,when the observed motion is greater than half of the radar wavelength,the phase will be wrapped.Therefore,the prerequisite for phase measurement using InSAR technology is to obtain the real phase from the wrapped phase.This process is called phase unwrapping.Unlike traditional InSAR technology,the phase unwrapping problem in MT-InSAR is defined on a sparse three-dimensional(3-D)grid,rather than on a regular two-dimensional(2-D)network.Therefore,this paper uses the phase closure information between redundant interferograms to study the phase unwrapping algorithm.The main research contents and results are as follows:(1)The basic principles and methods of several time series InSAR technologies and phase unwrapping are introduced in detail.And according to the dimensional characteristics of phase unwrapping,the phase unwrapping algorithms are classified into three categories:2-D phase unwrapping algorithm,1+2-D phase unwrapping algorithm and 3-D-phase unwrapping algorithm.(2)The 1+2-D phase unwrapping algorithm and the 3-D phase unwrapping algorithm are analyzed,and their problems in areas with drastic phase changes are summarized.Aiming at areas where the phase continuity assumption fails due to large deformation gradients,sudden terrain changes and system noise effects,a L~1-norm sparse 2-D phase unwrapping algorithm based on reliable redundant network is proposed.The redundant network is constructed by implementing the ambiguity detector on a distance-constrained free network.Experiment on real synthetic aperture radar interferograms covering urban tall buildings verified that the algorithm can obtain more reliable results in phase discontinuous regions than traditional PU algorithms.(3)Aiming at the problem that the traditional L~P-norm phase unwrapping algorithm cannot allocate the weights reasonably,a residual weighting method is proposed.Through simulation and real data experiments,it is shown that the unwrapping result obtained by this method is more accurate than the result obtained by the traditional weighting method,which proves that the weight distribution of this method is more reasonable than the other two methods.(4)Aiming at the problem that the traditional Delaunay triangulation network does not serve the phase unwrapping,a phase unwrapping network optimization method based on the simulated annealing algorithm is proposed.The simulation and real data experiments prove that the accuracy of the unwrapping results based on the optimized network is higher than that of the Delaunay triangulation.