不确定网络环境中任务卸载和资源分配联合优化方法  

Joint optimization method of task offloading and resourceallocation under uncertain wireless network

作　　者：王昭[1] 张承宇 左琳立 刘超超 WANG Zhao;ZHANG Chengyu;ZUO Linli;LIU Chaochao(No.10 Institute of CETC,Chengdu 610036,P.R.China;School of Communications and Information Engineering,Chongqing University of Posts and Telecommunications,Chongqing 400065,P.R.China;School of Communication Engineering,Chongqing Polytechnic University of Electronic Technology,Chongqing 401331,P.R.China)

机构地区：[1]中国电子科技集团公司第十研究所,成都610036 [2]重庆邮电大学通信与信息工程学院,重庆400065 [3]重庆电子科技职业大学通信工程学院,重庆401331

出　　处：《重庆邮电大学学报（自然科学版）》2024年第4期836-846,共11页Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition)

基　　金：国家自然科学基金项目(62071077,62301099);中国博士后科学基金项目(2023MD734137);重庆市自然科学基金创新发展联合基金项目(2022NSCQ-LZX0191)。

摘　　要：在不确定的无线网络环境中,由于任务到达、用户移动的随机性以及无线信道状态的时变性,导致出现不确定的任务云端排队时延、设备网络连接时间等网络环境特征,极大影响计算卸载效率和网络资源利用率。针对这一问题,建立了算网协同的动态任务卸载和资源联合分配模型,以最小化系统总能耗为目标,提出了基于随机模拟的任务卸载和算网资源联合分配多阶段随机规划优化算法(SS-MSSP),采用多阶段随机规划理论制定多阶段策略,并以后验算网资源分配的方式来追索补偿不确定网络环境的影响。仿真结果表明,在不确定的网络环境中,SS-MSSP算法保证了用户的计算时延需求,同时有效降低了系统能耗。In an uncertain wireless network environment,the randomness of task arrival,user movement,and the time-varying nature of wireless channel states lead to uncertain network environment characteristics such as cloud queuing delay and device network connection time,which greatly affect the efficiency of computing offloading and network resource utilization.To address this issue,this paper establishes a dynamic task offloading and edge resource adaptive allocation model for computing and network coordination.A multi-stage stochastic programming optimization algorithm(SS-MSSP)for task offloading and joint resource allocation based on stochastic simulation was proposed with the goal of minimizing the total energy consumption of the system.The multi-stage stochastic programming theory was used to formulate multi-stage strategies,and the posteriori computing network resource allocation method was used to compensate for the impact of uncertain network environments.Simulation results show that in an uncertain network environment,the SS-MSSP algorithm ensures the user’s computing delay requirements while effectively reducing system energy consumption.

关 键 词：移动边缘计算 不确定网络 任务卸载 资源分配 多阶段随机规划 

分 类 号：TN929.5[电子电信—通信与信息系统]