基于深度强化学习的实时视频边缘卸载策略  

Computation offloading decision in video edge computing based on deep reinforcement learning

作　　者：周陈静 骆淑云 ZHOU Chenjing;LUO Shuyun(School of computer science and technology(Artificial Intelligence),Zhejiang Sci-Tech University,Hangzhou 310018,China)

机构地区：[1]浙江理工大学计算机科学与技术(人工智能)学院,杭州310018

出　　处：《智能计算机与应用》2024年第8期32-39,共8页Intelligent Computer and Applications

基　　金：浙江省尖兵研发攻关计划项目(2023C01041)。

摘　　要：视频边缘计算架构将视频任务部分或全部卸载至靠近网络的边端进行处理,为实时视频处理提供了一个解决方案。为加速视频处理,本文搭建了一个视频边缘计算系统。针对具体的疵点检测视频任务,考虑视频处理的实时性和准确度要求,对计算卸载问题进行建模;为使卸载策略更好地适应动态的网络环境,将问题转换为马尔可夫决策过程,提出了一种基于深度强化学习和队列预测的视频计算卸载策略。实验结果表明,该卸载策略相较于基准策略能够在保证准确度的情况下,最大程度地降低系统时延。The video edge computing architecture provides a solution for real-time video processing by offloading partial or all of the video to the edge end close to the network for processing.A video edge computing system is built to accelerate video processing.The computational offloading problem is modelled for a specific fault detection video task,taking into account the real-time and accuracy requirements of video processing.In order to adapt to the dynamic network environment,the problem is transformed into a Markovian decision process and a deep reinforcement learning-based video computational offloading policy is proposed.The experimental results indicate that the offloading strategy is able to minimize system latency with guaranteed accuracy compared to other baseline strategy.

关 键 词：视频边缘计算 实时视频处理 深度强化学习 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TN929.5[自动化与计算机技术—控制科学与工程]