摘要
车载边缘计算(Vehicular Edge Computing, VEC)是移动边缘计算(Mobile Edge Computing, MEC)在车联网中的一个重要应用。在VEC中,请求服务的车辆可以通过付费的方式,将计算任务卸载到VEC服务器或者空闲计算资源丰富的服务车辆上,从而满足车辆任务对计算服务的需求。然而,对于VEC运营商来说,收益最大化是其追求的目标之一。由于系统中的计算需求和计算资源是动态变化的,因此如何在车辆协同场景下确定一个合理的定价策略是一个不容忽视的问题。针对该问题制定了一个动态定价策略,使VEC服务器和服务车辆的价格随着计算资源的供需关系而动态调整。基于此,设计了运营商收益最大化的车辆协同计算卸载方案,通过将时延约束下的VEC运营商收益最大化问题转化为多用户匹配问题,使用Kuhn-Munkres(KM)算法求得卸载结果。仿真实验表明,相比已有定价策略,该动态定价策略下VEC服务器和服务车辆的价格均可以根据计算资源供需关系动态调整,从而实现运营商收益最大化;相比已有卸载方案,该方案可以在满足任务时延约束的前提下提高运营商的收益。
Vehicular edge computing(VEC) is an important application of mobile edge computing(MEC) in Internet of vehicles.In VEC,to meet the computing requirement of task vehicles(TaV),TaV can pay to offload tasks to the VEC server or service vehicles(SeV)with abundant idle computing resources.For the VEC provider, one of its goals is to maximize revenue.Since the computing requirements and the computing resources of the system change dynamically, it is an important issue to design a reasonable pricing strategy in vehicle collaboration scenarios.To solve this problem, this paper designs a dynamic pricing strategy.In this strategy, service prices of the VEC server and SeV are adjusted dynamically according to the relationship of the supply and demand of computing resources.On this basis, a vehicle collaborative computation offloading scheme is designed to maximize provider’s revenue.By transforming the revenue maximization problem of VEC provider under the delay constraint into a multi-user matching problem, the offloading results are obtained using the Kuhn-Munkres(KM)algorithm.Simulation results show that compared to existing strategies, with this dynamic pricing strategy, the prices of the VEC server and SeV can be adjusted dynamically with the supply and demand of resources, so as to maximize the provider’s revenue.Compared to existing offloading schemes, this scheme can improve the provider’s revenue while meeting task delay.
作者
孙慧婷
范艳芳
马孟晓
陈若愚
蔡英
SUN Hui-ting;FAN Yan-fang;MA Meng-xiao;CHEN Ruo-yu;CAI Ying(Computer School,Beijing Information Science&Technology University,Beijing 100101,China)
出处
《计算机科学》
CSCD
北大核心
2022年第9期242-248,共7页
Computer Science
基金
国家自然科学基金(61672106)
北京市自然科学基金(L192023)
北京信息科技大学基金(2025028)
促进高校内涵发展-面向边缘计算的创新科研平台建设项目(2020KYNH105)
网络文化与数字传播北京市重点实验室开放课题
北京信息科技大学“勤信人才”培育计划(QXTCP C202111)。
关键词
车载边缘计算
计算卸载
协同计算
动态定价
Vehicular edge computing
Computation offloading
Collaborative computing
Dynamic pricing
