面向用户移动性的VLC辅助的UAV网络能效优化(特邀)OA

[Objective]In Visible Light Communication(VLC)-enabled Unmanned Aerial Vehicle(UAV)networks,UAVs ad-just their locations to keep Line-of-Sight(LoS)links with moving ground users.As a result,the limited onboard energy of a UAV must cover both communication tasks and flight propulsion.[Methods]To improve the Energy Efficiency(EE)of these networks,we propose an iterative trajectory optimization algorithm.Instead of solving a coupled global problem,we divide it into two alternating subproblems:optimal positioning and trajectory planning.First,we apply the Soft Actor-Critic(SAC)algorithm to jointly optimize the UAV's target position,user association,and power allocation.Once the target position is set,we model the flight trajectory to minimize propulsion energy as a convex problem,which can be solved based on standard optimization tools.[Results]Simulation results confirm the convergence of the SAC algorithm in our setup.We show the optimized UAV locations and user association schemes that lead to the highest network EE,along with the actual flight trajectories generated for dynamic user mobility scenarios.[Conclusion]Overall,our iterative framework successfully lowers the time-averaged energy consump-tion of the network,which directly extends the flight endurance of the UAVs.