Adaptive Weighting Scheme for Multi-Objective Optimization in Metasurface Antenna Design

We present a novel adaptive weighting scheme suited to the current-based optimization of metasurface antennas. The problem is formulated in terms of a weighted sum of individual objective functions corresponding to different constraints. A nonlinear conjugate gradient optimization algorithm is combined with an hyperplane adaptive weighting scheme to improve the convergence properties. The proposed approach is inspired by a geometrical interpretation of the properties of the Pareto front, and guarantees a balancing of the individual goals. The procedure has been applied to the design of broadside-radiating metasurface antenna working at 23 GHz, and demonstrated its effectiveness in improving both the speed of convergence and quality of the solution with respect to existing algorithms.