Intelligent Energy-Efficiency Trajectory Planning of Heterogeneous Air Base Stations for 6G Emergency Communication

Abstract

With the development of 6G, emergency communication services upgrade and the need for edge intelligence is increasing. However, today’s 6G emergency communication scenarios are changeable and complex, and existing methods cannot achieve large-scale and QoS guarantees. So we proposed a heterogeneous coverage compensation mechanism with static complete deployment and dynamic enhancement deployment. To enhance the capacity of hotpots determined by the clustering algorithm, we jointly optimize the UAV trajectory and users’ connectivity to maximize the UAV’s energy efficiency(EE). The optimized problem is NP-hard and is solved by deep reinforcement learning(DRL) algorithm. Simulation results show our method can significantly improve EE, which is much higher than Q-learning, PSO, and GA.