Trajectory Design and Bandwidth Assignment for UAVs-enabled Communication Network with Multi - Agent Deep Reinforcement Learning
Authors:
- 1 September 2021
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Abstract
Unmanned aerial vehicle (UAV) is considered as a promising technique to enhance future wireless mobile communication. In this paper, the UAVs serve as aerial base stations (UBSs) to provide reliable service for terminal users (TUs). We construct the physical model and air-ground link model to formulate an optimization problem, aiming to maximize the total channel capacity of the network. To address this non-convex problem, we proposed a multi-agent deep reinforcement learning algorithm (MADRL) based on deep deterministic policy gradient (DDPG) to optimize the trajectory design and bandwidth assignment dynamically. Proper reward function is well designed to guide the agents to exploit the potential of the aerial network. Simulations demonstrate that our scheme can adapt the rapidly changing environment and obtains 29.49% performance improvement comparing to the baseline scheme.Keywords
This publication has 14 references indexed in Scilit:
- Resource Allocation Optimization of UAVs-Enabled Air-Ground Collaborative Emergency Network in Disaster AreaInternational Journal of Performability Engineering, 2019
- Beyond 5G With UAVs: Foundations of a 3D Wireless Cellular NetworkIEEE Transactions on Wireless Communications, 2018
- Energy-Efficient UAV Control for Effective and Fair Communication Coverage: A Deep Reinforcement Learning ApproachIEEE Journal on Selected Areas in Communications, 2018
- Joint Trajectory and Communication Design for Multi-UAV Enabled Wireless NetworksIEEE Transactions on Wireless Communications, 2018
- Mastering the game of Go without human knowledgeNature, 2017
- Energy-Efficient UAV Communication With Trajectory OptimizationIEEE Transactions on Wireless Communications, 2017
- Echo State Networks for Self-Organizing Resource Allocation in LTE-U With Uplink–Downlink DecouplingIEEE Transactions on Wireless Communications, 2016
- Efficient 3-D placement of an aerial base station in next generation cellular networks2016
- Human-level control through deep reinforcement learningNature, 2015
- Optimal LAP Altitude for Maximum CoverageIEEE Wireless Communications Letters, 2014