Research Article
Trajectory Optimization and Power Control Of UAV-Assisted Mobile Edge Devices
@INPROCEEDINGS{10.1007/978-3-031-60347-1_32, author={Yunfeng Xia and Qingling Liu and Shihao Wang and Kuixian Li}, title={Trajectory Optimization and Power Control Of UAV-Assisted Mobile Edge Devices}, proceedings={Mobile Multimedia Communications. 16th EAI International Conference, MobiMedia 2023, Guilin, China, July 22-24, 2023, Proceedings}, proceedings_a={MOBIMEDIA}, year={2024}, month={10}, keywords={Mobile edge server Trajectory optimization Power control Deep reinforcement learning Multi-agents}, doi={10.1007/978-3-031-60347-1_32} }- Yunfeng Xia
Qingling Liu
Shihao Wang
Kuixian Li
Year: 2024
Trajectory Optimization and Power Control Of UAV-Assisted Mobile Edge Devices
MOBIMEDIA
Springer
DOI: 10.1007/978-3-031-60347-1_32
Abstract
Mobile edge computing (MEC) is an important aspect of 5G networks, and the deployment of mobile edge devices on unmanned aerial vehicles (UAVs) is becoming an increasingly popular trend in the future. However, when the communication bandwidth resources are limited, the co-frequency link interference in the UAV-assisted MEC may lead to poor communication quality. Thus, this paper proposes a trajectory planning and power allocation scheme based on multi-agent deep reinforcement learning (DRL). To begin, we design a joint optimization goal, which aims to minimize link interference while simultaneously maximizing communication link throughput. Next, we propose an improved DRL algorithm for a multi-agent scenario that enables autonomous trajectory planning and power control of UAVs, ensuring the reasonable allocation of resources. Experimental results demonstrate that the proposed algorithm has better convergence and greater revenue compared to other benchmark algorithms. Overall, this paper presents a promising approach for improving the performance of UAV-assisted MEC through intelligent decision-making and resource allocation.
