Reduced-cost design closure of antennas by means of gradient search with restricted sensitivity update

Design closure, i.e., adjustment of geometry parameters to boost the performance, is a challenging stage of antenna design process. Given complexity of contemporary structures, reliable parameter tuning requires numerical optimization and can be executed using local algorithms. Yet, EM-driven optimization is a computationally expensive endeavour and reducing its cost is highly desirable. In this paper, a modification of the trust-region gradient search algorithm is proposed for accelerated optimization of antenna structures. The algorithm is based on sparse updates of antenna sensitivities involving various methods that include the Broyden formula used for selected parameters, as well as dimensionality- and convergence-dependent acceptance thresholds which enable additional speedup, and make the procedure easy to tune for various numbers of antenna parameters. Comprehensive verification executed for a set of benchmark antennas delivers consistent results and considerable cost reduction of up to 60 percent with respect to the reference algorithm. Experimental validation is also provided.