移动边缘计算网络中的资源分配与定价  

Resource Allocation and Pricing in Mobile Edge Computing Networks

作　　者：吕晓东 邢焕来[2] 宋富洪 王心汉 LYU Xiao-Dong;XING Huan-Lai;SONG Fu-Hong;WANG Xin-Han(School of Information Science and Technology,Southwest Jiaotong University,Chengdu 610031,China;School of Computing and Artificial Intelligence,Southwest Jiaotong University,Chengdu 610031,China)

机构地区：[1]西南交通大学信息科学与技术学院,成都610031 [2]西南交通大学计算机与人工智能学院,成都610031

出　　处：《计算机系统应用》2022年第10期99-107,共9页Computer Systems & Applications

摘　　要：移动边缘计算(mobile edge computing,MEC)使移动设备(mobile device,MD)能够将任务或应用程序卸载到MEC服务器上进行处理.由于MEC服务器在处理外部任务时消耗本地资源,因此建立一个向MD收费以奖励MEC服务器的多资源定价机制非常重要.现有的定价机制依赖于中介机构的静态定价,任务的高度动态特性使得实现边缘云计算资源的有效利用极为困难.为了解决这个问题,我们提出了一个基于Stackelberg博弈的框架,其中MEC服务器和一个聚合平台(aggregation platform,AP)充当跟随者和领导者.我们将多重资源分配和定价问题分解为一组子问题,其中每个子问题只考虑一种资源类型.首先,通过MEC服务器宣布的单价,AP通过解决一个凸优化问题来计算MD从MEC服务器购买的资源数量.然后,MEC服务器计算其交易记录,并根据多智能体近端策略优化(multi-agent proximal policy optimization,MAPPO)算法迭代调整其定价策略.仿真结果表明,MAPPO在收益和福利方面优于许多先进的深度强化学习算法.Mobile edge computing(MEC)enables mobile devices(MDs)to offload tasks or applications to MEC servers for processing.As a MEC server consumes local resources when processing external tasks,it is important to build a multiresource pricing mechanism that charges MDs to reward MEC servers.Existing pricing mechanisms rely on the static pricing of intermediaries.The highly dynamic nature of tasks makes it extremely difficult to effectively utilize edge-cloud computing resources.To address this problem,we propose a Stackelberg game-based framework in which MEC servers and an aggregation platform(AP)act as followers and the leader,respectively.We decompose the multi-resource allocation and pricing problem into a set of subproblems,with each subproblem only considering a single resource type.First,with the unit prices announced by MEC servers,the AP calculates the quantity of resources for each MD to purchase from each MEC server by solving a convex optimization problem.Then,each MEC server calculates its trading records and iteratively adjusts its pricing strategy with a multi-agent proximal policy optimization(MAPPO)algorithm.The simulation results show that MAPPO outperforms a number of state-of-the-art deep reinforcement learning algorithms in terms of payoff and welfare.

关 键 词：移动边缘计算(MEC) 资源定价 博弈论 深度强化学习 资源分配 

分 类 号：TN929.5[电子电信—通信与信息系统]