Microwave imaging method using a simulated annealing approach

A method of microwave imaging in the space domain which is based on a simulated annealing approach is presented. Markov random fields are used to model the distributions of dielectric features, and a stochastic relaxation algorithm is developed to solve the imaging problem. Computationally heavy inversions of large matrices are avoided. Numerical results confirming the capabilities of the proposed method are reported. They seem better than those obtained by other imaging techniques in the space domain.<
>