摘要
对车联网中多个车辆用户设备(Vehicle User Equipment,VUE)卸载过程中的资源分配问题进行研究,提出一种时延感知的计算卸载和资源分配策略。采用支持向量机将卸载任务根据时延和能耗的要求选择移动边缘计算(Mobile Edge Computing,MEC)服务器辅助卸载和VUE辅助卸载两种处理模式,考虑MEC服务器的计算资源分配以及车到车链路复用蜂窝链路的干扰问题,建立最小化总成本的优化问题。最后,使用深度竞争双Q网络算法完成不同处理模式下相应的资源分配过程。为了验证所提策略的有效性,将其与基于深度Q网络算法的资源分配策略、基于正交多址接入的资源分配策略和随机资源分配策略等3种策略相比。验证结果表明,所提策略可以在最大时延限制内有效降低卸载总成本。
The resource allocation problem in the offloading process of multiple vehicle user equipment(VUE)in the Internet of Vehicles is studied,and a delay-aware computation offloading and resource allocation strategy is proposed.Support vector machine is used to select two processing modes of mobile edge computing(MEC)server-assisted offloading and VUE-assisted offloading for the offloading task according to the requirements of delay and energy consumption,and to establish the optimization problem of minimizing the total cost by considering the interference of the vehicle-to-vehicle link multiplexing cellular link and the computational resource allocation of the MEC server.Finally,the dueling double deep Q-network algorithm is used to complete the corresponding resource allocation process under different processing modes.To verify the effectiveness of the proposed strategy,it is compared with three strategies,including the deep Q-network algorithm-based resource allocation strategy,orthogonal multiple access-based resource allocation strategy,and random resource allocation strategy.The validation results show that the proposed strategy can effectively reduce the total offloading cost within the maximum delay limit.
作者
江帆
李妍
宋琦琳
JIANG Fan;LI Yan;SONG Qilin(School of Communications and Information Engineering,Xi an University of Posts and Telecommunications,Xi an 710121,China;Shaanxi Key Laboratory of Information Communication Network and Security,Xi an University of Posts and Telecommunications,Xi an 710121,China)
出处
《西安邮电大学学报》
2023年第1期1-8,共8页
Journal of Xi’an University of Posts and Telecommunications
基金
国家自然科学基金项目(62071377,62201456)
陕西省自然科学基金项目(2021JM-465)
陕西省重点产业创新链项目(2020ZDLGY02-06)。
关键词
车联网
移动边缘计算
计算卸载
资源分配
支持向量机
深度竞争双Q网络
Internet of Vehicles
mobile edge computing
computation offloading
resource allocation
support vector machine
dueling double deep Q-network
