Design of 5G Dual-Antenna Passive Repeater Based On Machine Learning

In 5G communications, small cells are one of the main approaches to achieve data diversion and improve network capacity. The problem of blind area is partially solved by this way, because the distances between small base stations and users are cut short. However, the intensive deployment of small base stations will bring about complex disturbance and a large amount of energy consumption. To overcome this challenge, we propose a new approach of dual-antenna passive repeater, which consists of a four-element patch antenna array, a feeding network and an improved planar Yagi-Uda antenna with added parasitic patches. It can be used in cooperation with small base stations to replace the function of the small base stations in a certain point, change the beam pointing, and achieve wide-angle scattering to realize the blind area signal coverage. The genetic algorithm which is a branch of machine learning is used to optimize the antenna parameters. Simulation results show that our proposed passive repeater can effectively reduce the path loss and improve the signal power of the receiving end.