Design of Multibeam Dielectric Lens Antennas by Multiobjective Optimization

We propose an effective method for designing multibeam dielectric lens antennas. A genetic algorithm (GA) with multiobjective optimization is adopted to balance gain against sidelobe level. The lens shapes and the position of each feed are associated with chromosomes. The gain and sidelobe level are evaluated by a Pareto ranking method. The method yields the distribution of the objective function values and the corresponding antenna structures.