Parameter optimization of energy-efficient antenna system using period-based memetic algorithm

Antenna systems play a key role both in today's 5G or future 6G communication networks because they can convert electrical energy directly into electromagnetic waves. However, due to the imprecise antenna models, electrical energy loss is inevitable and enormous in the whole network. Therefore, the energy efficiency optimization of antenna models becomes extremely urgent by accurately estimating their parameters. In this paper, to reduce the energy loss in transmission, a novel period-based memetic algorithm (MA) framework is developed to improve the energy-saving efficiency of antenna models. A popular neighborhood field search (NFS) and a state-of-the-art differential evolution (DE), as a local optimizer and a global optimizer, respectively, are embedded into the MA framework for an instantiation, referred to as MDE-NFS. In the proposed MDE-NFS algorithm, a periodic switching-based scheme is studied to strike a balance between global exploration and local exploitation. To verify the performance of MDE-NFS, it is applied to parameter optimization of two different antenna models, including the microstrip antenna model and the Yagi–Uda antenna model. Moreover, MDE-NFS is also specifically compared with other algorithms on several numerical test sets. The comprehensively experimental results demonstrate that the MDE-NFS is promising to be a candidate parameter optimization approach to design energy-efficient antenna models.