Parameter Selection Criteria for Tomo-SAR Focusing

The synthetic aperture radar (SAR) tomography (TomoSAR) inverse problem is commonly tackled in the context of direction-of-arrival estimation theory. latter allows achieving super resolution, along with ambiguity levels reduction, thanks to use parametric focusing methods, as multiple signal classification (MUSIC) and statistical regularization techniques, like maximum-likelihood-inspired adaptive robust iterative approach (MARIA). Nevertheless, order correctly suit considered model, MUSIC most approaches require an appropriate setting involved parameters. In both cases, accuracy retrieved solutions depends on right selection assigned values. Thus, aim dealing such issue, this article addresses several parameter strategies, adapted specifically TomoSAR scenario. Parametric techniques solve a different manner methods do, hence, each demands methodologies for proper their Consequently, we refer Kullback-Leibler information criterion model MUSIC, whereas rather explore Morozov's discrepancy principle, L-Curve, Stein's unbiased risk estimate, generalized cross-validation choose After incorporation these criteria MARIA, respectively, capabilities are first analyzed through simulations, later on, utilizing real data acquired from urban area.