Robust antenna design through a hybrid inversion strategy combining interval analysis and nature-inspired optimization

The problem of designing robust antenna configurations is cast as the solution of an inverse problem that is addressed by means of a hybrid approach based on Interval Analysis (IA) and Particle Swarm Optimization (PSO). The admissible tolerance error on the surface of reflector antennas or on the element excitations of antenna array is maximized while guaranteeing the upper and lower pattern bounds, including all power patterns potentially generated by the antenna, satisfying user-defined mask constraints. The pattern bounds are analytically computed through the IA for each trial antenna configuration defined by means of the PSO. Representative numerical examples are shown to validate the effectiveness of the proposed approach.