Modelling and Analysis Tools for Interferometric Sar Observations

A quantitative deformation monitoring using the differential interferometric SAR (DInSAR) technique may be achieved when multiple DInSAR observations and suitable modelling and analysis tools are employed. The paper begins with a description of the main characteristics of the DInSAR data. Then, it discusses a new modelling and filtering strategy, which takes advantage of the specific properties of the DInSAR observations. The core of the procedure is the least squares collocation filtering and prediction, which exploits the correlation properties of the DInSAR data. The proposed procedure was tested on simulated DInSAR data that reproduce the characteristics of a small scale and slow deformation rate subsidence, and that include the main components of the interferometric data: the atmospheric contribution, the phase noise component, and the outliers due to the unwrapping related errors.