A multi-temporal DInSAR-based method for the assessment of the 3D rigid motion of buildings and corresponding uncertainties

Abstract

This paper investigates the performances of the multi-temporal DInSAR technique for the structural monitoring of single buildings. The application of the multi-temporal DInSAR technique at building or infrastructure scale is enabled by the high imaging capability of the current generation of SAR constellations. Within this framework, the paper proposes an analytical procedure for the reconstruction of the 3D rigid motion of single buildings based on SAR measurements. To successfully reconstruct the structural displacement, measurements from dual orbits are required by the method. The proposed procedure is tested through numerical analyses accounting for uncertainties affecting SAR data, both in the displacement measurements and in the positioning of the main persistent scatterers. Numerical simulation results allow for the quantification of uncertainties in the estimation of the motion parameters. An analytical model is also proposed to estimate the motion parameter uncertainties depending only on the building geometry, the satellite orbit inclinations and the number of persistent scatterers, with no need to know their exact positions. The main benefit of the proposed analytical model is that it can be adopted in real applications to a priori evaluate the precision by which the motion components can be estimated in order to assess the reliability of the structural monitoring based on SAR interferometry. A very good match between numerical and analytical motion parameter uncertainties is achieved. Finally, the proposed method to reconstruct the 3D rigid motion and assess the corresponding uncertainties is applied to a real case study in Rome (Italy).