Multi-UAVs Collaboration System based on Machine Learning for Throughput Maximization

Due to commercialization of the 5G network, many base stations need to enhance a reliable communication quality. Thus, many studies have still worked to provide mobility and economic benefits to the VAVs-Base Station (VAVs-BS) on behalf of ground base stations. In this paper, we propose a system to find a location where multiple users can have an optimal service throughput by considering users' requirements in Multi-VAVs communication. Based on the Air-To-Ground Path Loss Model, the virtual communication environment is established and Airtime Fairness is applied for equitable channel usage time distribution according to user requirements. Thus, we apply a collaborative algorithm with modified K-means that can distribute users to each VAV and solve communication overload problems. In addition, the Proximal Policy Optimization (PPO) algorithm is applied to set an optimal location with the maximum throughput. As a result, the proposed systems allow the Multi-VAVs to be in the locations with high service throughput for users with different demands.