| Title |
Deep Reinforcement Learning-based Energy Consumption Optimization Technology in 5G Drone-MEC Network |
| Authors |
고윤영(Yunyeong Goh) ; 오신혁(Shinhyeok Oh) ; 정종문(Jong-Moon Chung) |
| DOI |
https://doi.org/10.5573/ieie.2025.62.1.22 |
| Keywords |
5G; Drone; Deep reinforcement learning; Multi-access edge computing |
| Abstract |
Through images and video data collected from sky, drones can provide object recognition and traffic monitoring services. In order for drones to quickly provide 5G services, fast task processing speed is required, and energy consumption must be minimized due to the limited battery capacity. To reduce energy consumption, Multi-access Edge Computing (MEC) technology can be used to offload tasks to a server. However, task offloading causes transmission delay, and it is difficult to determine the optimal offloading ratio while satisfying several constraints. Therefore, in this paper, a DRL (Deep Reinforcement Learning)-based optimization technology is developed to minimize energy consumption in the drone-MEC network. Through simulation, it is shown that the proposed model optimally selects the offloading ratio and the transmission power value of multi-RAT (Radio Access Technology), resulting in excellent energy consumption performance. |