Research Of PS-InSAR Techology On Monitoring Surface Deformation In Xi'an

As we known,satellite synthetic aperture radar interferometry(InSAR) as a romising remote sensing technique,has been developed rapidly in recent twenty years,which has advantages of all-weather,large-scale and high accuracy.It's a new economy technology in deformation disaster monitor.However,It is very hard to form interferogram for limited factors such as spatio-temporal decorrelation and stmospheric effect.PS-InSAR technique has the advantage to overcome the limited factors.The permanent scatters(PS),which can keep phase stable in the temporal series,can be used to get surface subsidence by analyzing the times-series differential phase variation.In this paper,theory of InSAR are first introduced,and the components of interfergram phases are analyzed.The methods of InSAR and D-InSAR were introduced.We constructed function model of differential phases and analyzed the factors effecting phase quality.In order to improve the reliability and accuracy of PS identification,a dual-thresholding algorithm based correlation-coefficient thresholding and amplitude dispersion index thresholding is proposed to improve the quality of PS points detection.This method consider the phase stability and the high SNR of PS points.The testing result of 17 interferometric pair over xi'an showed that this method can identify PS points effectively.Compared to other unwrapping methods,we choice LAMBDA method.We constructed function model between time series spational differential phases and analyzed the process of phase unwrapping.Using the LAMBDA method,deformation rates are sloved by a spatial adjustment.And the QUAD are used to detect gross errors.At last,we chose Matlab program to process the subsidence of xi'an.Time series interferogram phases are generated by 17 SAR images between 2003 and 2007.We selected PS points using dual-thresholding algorithm and constructed model using Delaunay triangulation.The maximam subsidence rate which is -7cm / a can be computed using LAMBDA method.With the comparison of GPS and InSAR monitoring data in the same period,it confirmed the effective of PS-InSAR in monitoring surface deformation.In this paper,the theoretical model and algorithm of the PS-InSAR technology is achieved.And a completely computational procedures is independently developed.We can separate deformation information and the DEM error.The first time we use PS-InSAR technology in xi'an surface deformation monitoring and make optimistic results.The results confirmed that the algorithm and program are efficient and reliable.