Estimating Scatterer Positions using Sparse Approximation

Other conference contribution, 2016

We present a convex optimization method for a class of inverse scattering problems. The method is based on three steps: (i) compute a database of scattering data for the measurement situations of interest; (ii) find a sparse approximation of a measured response in terms of the database; and (iii) estimate a representative description from the sparse approximation as a weighted average. We use the method to estimate the position of multiple scatters inside a microwave measurement system.