• edge-preserving regularization
• evolutionary algorithms
• range profile reconstruction
• ___ microwave tomography

In the present work we have presented a reliable and efficient algorithm for the data inversion, which is based on a fully nonlinear data model in conjunction with an optimization technique. The reconstruction of the permittivity range profile has been tested both on
synthetic and real data to validate the electromagnetic code as well as to assess the accuracy and efficiency of the reconstruction procedure. We have studied the resolution of the algorithm and its robustness to the noise, demonstrating the ability of our procedure to be able to recognize the presence of high discontinuities even independently from the discretization fixed by the user.
As a part of the ongoing improvement of the presented method, we have addressed the implementation of a new optimization algorithm, namely the particle swarm optimization, which has been customized and enhanced for our purposes.
Finally, a detailed description of a fast and efficient procedure to evaluate the green’s function for a multilayered medium has been given. This is the groundwork useful for the next step toward a more reliable and versatile forward solver to be implemented in the inversion procedure.