An Optimal Approach to the Inverse Problem

Abstract

We describe how remote sensing problems can be reformulated within the framework of optimization theory. This reformulation allows any prior knowledge about the solution to be naturally incorporated into the solution process. The inversion problem then reduces to a search for the global extremum in the possible presence of local extrema. Two algorithms are presented that can be used to solve this global optimization problem, and their application to the helioseismology inverse problem is detailed.