On the exploitation of the a-priori information through the Bayesian compressive sensing for microwave imaging

A set of computationally efficient inverse scattering techniques that exploit the `a-priori´ information on the sparseness of the unknown scatterers is presented. Towards this end, the Contrast Source formulation of the inverse scattering problem is presented within the Bayesian Compressive Sampling (BCS) framework and successively solved through a Relevance Vector Machine. Some illustrative examples are provided to show the features and potentialities of the approach both when dealing with TE and with TM scattering data.