Stochastic Systems Group
Home Research Group Members Programs  
Demos Calendar Publications Mission Statement Alumni

SSG Seminar Abstract


Anisotropy Characterization in Synthetic Aperture Radar using Sparse Signal Representation

Kush Varshney
SSG, LIDS, MIT


We consider the problem of jointly forming images and characterizing anisotropy from wide-angle synthetic aperture radar (SAR) measurements.  Conventional SAR image formation techniques assume isotropic scattering, which is not valid with wide-angle apertures.  We present a method based on a sparse representation of aspect-dependent scattering with an overcomplete basis composed of basis vectors with varying levels of angular persistence.  Solved as an inverse problem, the result is a complex-valued, aspect-dependent response for each spatial location in a scene.  Our non-parametric approach considers all point scatterers in a scene jointly.  The choice of the overcomplete basis set incorporates prior knowledge of aspect-dependent scattering, but the method is flexible enough to admit solutions that may not match a family of parametric functions.  We enforce sparsity through regularization based on the lp quasi-norm, p<1.  This formulation leads to an optimization problem that is solved through robust half-quadratic methods.  We also develop a graph-structured interpretation of the overcomplete basis leading towards approximate algorithms using hill-climbing search with appropriate stopping conditions and search heuristics.  We present experimental results on synthetic scenes and on realistic radar measurements of a man-made vehicle.



Problems with this site should be emailed to jonesb@mit.edu