The genetic algorithm is used to reconstruct the shape what the initial estimate is [9]. In Section II, a theoretical formuof a perfectly conducting cylinder illuminated by transverse electric lation for the inverse scattering is presented. The general princi-(TE) waves. A cylinder of unknown shape scatters the incident TE ples of genetic algorithms and the way we applied them to the wave in a free space and the scattered field is recorded outside. inverse problem are described. Numerical results for perfectly Based on the boundary condition and the measured scattered field, conducting objects are given in Section III. Finally, some conclua set of nonlinear integral equations is derived and the imaging probsions are drawn in Section IV. lem is reformulated into an optimization problem. The genetic algorithm is then employed to find out the global extreme solution of II. THEORETICAL FORMULATION the cost function. Numerical results demonstrated that the genetic algorithm can tackle the inverse problem of a larger scatterer. Even A. Imaging Problem. The geometry of a two-dimensional imwhen the electrical dimension of the scatterer exceeds one wave-
aging problem is shown in Figure 1. A perfectly conducting length and the initial guess is far from the exact one, good reconstruccylinder is illuminated by an incident plane wave with a magnetic tion was obtained. In such a case, gradient-based methods often get field vector parallel to the z axis (i.e., TE polarization). The stuck in a local extreme. In addition, the effect of Gaussian noise on cylinder is of an infinite extent in the z direction, and its crossthe reconstruction results is investigated.