Concurrent estimation of space and frequency variation for dielectrics: a microwave tomography system for process sensing applications

Paper in proceeding, 2011

In many powder processing operations, the effective complex permittivity of the material under processing yields important information concerning the state of the process. This complex permittivity varies in general with respect to both space and frequency. For an accurate estimation of the process material parameters, it is important to be able to incorporate these dependencies in the estimation algorithm. Here, we present a rather general type of parameterization that is tailored for this type of situation. The parameterization is described in terms of a parameterization mesh, which makes it feasible to refine the computational mesh without influencing the number of degrees of freedom in the estimation problem. A test case demonstrates that our approach can yield accurate reconstructions.