Sparsity based TomoSAR combining CS and GLRT

In this paper a new approach to TomoSAR imaging is presented. It is based on the joint use of a Constant False Alarm Rate (CFAR) detection approach of multiple targets and of Compressive Sampling (CS) tomographic reconstructions. CS is widely used to recover a sparse signal but suffers from the presence of outliers. The proposed method consists in applying a Generalized Likelihood Ratio Test (GLRT) exploiting the CS reconstruction in order to detect and accurately localize single and double scatterers with a given false alarm probability, avoiding outliers and artefacts.