The Grey Wolf Optimizer for Antenna Optimization Designs: Continuous, binary, single-objective, and multiobjective implementations

The grey wolf optimizer (GWO) is a newly invented metaheuristic that simulates the hunting process of grey wolves in nature. As a robust optimization technique, the GWO engine has the capacity of handling antenna optimization problems with both continuous and binary variables and single and multiple objectives. In this article, the GWO and its binary (BGWO) version are introduced first. Their multiobjective versions, i.e., (MOGWO) and (MOBGWO), respectively, follow. To show the versatility of the GWO engine, some typical antenna optimization design problems are considered. In particular, a low-sidelobe sparse linear array and a high-directivity Yagi–Uda antenna are optimized by continuous GWO (CGWO); a thinned planar array is designed by a BGWO for sidelobe suppression in the two principal planes. To evaluate the performance of the GWO engine, comparative studies of the GWO with two popular optimization algorithms, i.e., a genetic algorithm (GA) and particle swarm optimization (PSO), are presented. It turns out that the GWO can, in most cases, outperform a GA and PSO. Further, these examples are expanded to consider more than one objective, and multiobjective versions of CGWO and BGWO, respectively, are employed to obtain the Pareto fronts, which clearly show the best tradeoffs that can be made.