无人机辅助移动边缘计算中的视频任务卸载算法  

Video task offloading algorithm in UAV-assisted mobile edge computing

作　　者：胡文杰 卢先领[1] Hu Wenjie;Lu Xianling(School of Internet of Things Engineering,Jiangnan University,Wuxi Jiangsu 214122,China)

机构地区：[1]江南大学物联网工程学院,江苏无锡214122

出　　处：《计算机应用研究》2025年第4期1128-1134,共7页Application Research of Computers

基　　金：国家自然科学基金资助项目(61773181)。

摘　　要：无人机(unmanned aerial vehicles,UAV)在未来通信网络中的集成备受关注,在军事侦察、火灾监控等诸多应用中发挥着至关重要的作用。针对此类场景中对视频传输低时延和高体验质量(quality of experience,QoE)的需求,提出了一种无人机辅助的移动边缘计算(mobile edge computing,MEC)视频任务卸载算法。首先,为满足低时延需求,无人机作为MEC服务器对视频数据进行转码,并作为中继将视频数据转发到地面基站。其次,通过联合优化设备关联、传输功率、转码策略以及无人机飞行轨迹,构建最大化系统效益的模型。最后,提出一种基于柔性动作-评价(soft actor-critic,SAC)的深度强化学习算法作出联合决策。仿真实验结果表明,与其他基线算法相比,所提算法能有效降低系统时延并保证视频任务品质,并具有良好的收敛性。The integration of unmanned aerial vehicles(UAV)in future communication networks has received great attention,and it plays an essential role in many applications,such as military reconnaissance,fire monitoring,etc.Aiming at the requirement of low latency and high quality of experience(QoE)of video transmission in these scenarios,this paper proposed UAV-assisted mobile edge computing(MEC)video task offloading algorithm.Firstly,to meet the low latency requirements,UAV acted as the MEC server to transcode videos and served as relays to forward the transcoded videos to ground base stations.Additionally,it constructed a system utility maximization model by jointly optimizing device association,transmission power,transcoding strategy,and UAV flight trajectory.It solved the optimization problem using deep reinforcement learning,and proposed a soft actor-critic(SAC)-based algorithm.Simulation results demonstrate that the SAC-based optimization algorithm effectively reduces system latency while ensuring video task quality,while exhibiting good convergence properties.

关 键 词：无人机 移动边缘计算 视频传输 任务卸载 深度强化学习 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]