A Kalman Smoothing Approach for Surface Deformation Monitoring in Differential SAR Interferometry

This paper presents a Kalman smoothing approach for estimating the temporal evolution of terrain deformations, using a chronologically ordered sequence of differential synthetic aperture radar interferograms spanning a long time interval. The proposed method assumes that the interferograms are formed with respect to the same master image and that the data pairs used to generate the interferograms can have orbital separations (i. e. baselines) up to the decorrelation baseline or even larger. The algorithm presented next applies to every highly coherent pixel (known as permanent scatterers) in the image and it is independent of the interferometric processing of the SAR images; so it is essentially a post-processing step. Since it's supposed that all the SAR images are available for batch processing, the proposed algorithm is naturally a smoother. The Kalman smoother reported thereafter can be exploited successfully to estimate the line-of-sight surface deformation of a vast area. This technique has been tested on simulated deformation data and it has demonstrated an excellent capability to track the ground deformation dynamics with a high accuracy.