The STUN algorithm for persistent scatterer interferometry

Abstract

The Spatio-Temporal Unwrapping Network (STUN) is a new algorithm to estimate displacement and topography at Persistent Scatterer (PS) points using an interferometric single-master data stack. The STUN algorithm provides a robust method to explicitly unwrap the interferometric phase using the temporal and spatial correlations of the observed phase. Moreover, it uses alternative hypothesis tests and the integer Least-Squares estimator to optimally estimate the parameters. Key features are the ability (i) to model the displacement using a linear combination of basefunctions; (ii) to estimate the precision of each SLC image using a novel stochastic model; (iii) to obtain a description of the precision of the estimated displacements. This paper describes the STUN algorithm in detail, and demonstrates its practical application using data from a rural area nearby Marseille, France, which suffers from subsidence due to mining activities.