Research On Methods Of High-Precision Three-Dimensional Deformation Resolving Based On Multi-Track InSAR Data

Small Baseline Subset Interferometric Synthetic Aperture Radar(SBAS-InSAR)is developed based on Differential Interferometric Synthetic Aperture Radar(D-InSAR).SBAS-InSAR technology not only inherits the advantages of D-InSAR technology,such as large-scale,high efficiency,and all-weather,but also can reduce the influence of decoherence noise and atmospheric delay,and obtain high-precision and long-term onedimensional Line of Sight(LOS)deformation.It is an important technical means for mining subsidence monitoring.However,in the practical application of solving threedimensional deformation using multi-orbit LOS data,the solution results are greatly affected by the inconsistency of the revisit period and the ill-posedness of the solution model,especially since the obtained north-south deformation results are far from the true value.Because of the above problems,this thesis summarizes the research status of threedimensional deformation resolving method based on multi-track InSAR,and then proposes to apply interpolation algorithm and biased estimation theory to the field of mining area deformation monitoring to reconstruct high-precision three-dimensional deformation in the mining area.The main research contents and innovations are as follows :(1)SBAS-InSAR technology is used to process the Sentinel-1A data of 35 ascending T62,33 descending T84 and 35 descending T157 to obtain multi-orbit LOS data.Aiming at the problem of missing and time asynchrony of multi-orbit time series InSAR data,a cubic interpolation algorithm is used to process the data,and 102 three-orbit data with the same revisit period are obtained.Compared with the same period of Global Navigation Satellite System(GNSS)data,the deformation magnitude and deformation trend of the157 orbit interpolation results are basically the same.The maximum root mean square error of the sampling points in the subsidence area is 4.65 mm.The accuracy of the cumulative deformation results of multi-orbit time series is high,which can be used to extract threedimensional deformation with higher accuracy.(2)Based on the error propagation process of the solution model,an objective function constraint model(TLEP-InSAR)for inversion of high-precision threedimensional deformation is proposed,and the three-dimensional deformation field of the Wuhai mining area is obtained by using three LOS measurements.In order to verify the effectiveness of the method,three schemes are designed to compare and analyze the GNSS data of sampling points.The results show that the TLEP-InSAR scheme is superior to the other two methods.The minimum root mean square error of three-dimensional cumulative deformation is 3.07 mm in the vertical direction,2.78 mm in the east-west direction,and5.89 mm in the north-south direction.(3)Based on the ill-posed problem of the joint solution model based on multi-track InSAR,this thesis gives the ill-posed degree of the model under different satellite parameters and proposes a three-dimensional deformation solution model(TR-InSAR)based on regularization method.The model obtains high-precision three-dimensional deformation information by weakening the pathological degree of the solution model.The results show that compared with the MSBAS-3D method,the north-south accuracy is improved by about 94 %.Compared with the TLEP-InSAR model,the accuracy of smallscale north-south deformation points is improved by about 50 %.Through simulation experiments,the solution ability of the TR-InSAR model and TLEP-InSAR model in illposed problems is verified.The results show that the TR-InSAR solution model has better anti-interference ability.