Toward Energy-Efficient UAV-Assisted Wireless Networks Using an Artificial Intelligence Approach
Shu Fu, Meng Zhang, Min Liu, Chen Chen, F. Richard Yu
Abstract
This article studies the application of artificial intelligence (AI) approach in UAV-assisted wireless networks to cope with a large number of parameters impacting energy-efficiency in the sixth generation wireless network. In order to improve the energy efficiency for UAV-assisted wireless networks, we focus on the following three aspects: the UAVs trajectory planning; caching, computing, and communication resource allocation of UAVs; and 3D hovering location decision of UAVs. We discuss each aspect and reveal the corresponding optimization problem of energy efficiency. We also explore several promising deep-learning-based AI methods, which include pointer network, federated deep learning, and multi-agent deep deterministic policy gradient, to solve these optimization problems. Through case studies, we verify the superiority of the proposed AI methods to save UAVs' energy and decrease the system delay.