一种能效优先的物联网任务协同迁移策略
|
摘要
移动边缘计算通过在数据源端执行通信和计算操作,缩减了物联网业务的传输和处理时延。然而,针对大量的物联网设备连接数,海量碎片化的数据同时汇聚在边缘计算平台,会显著地增加前传链路的流量负载和边缘服务器的计算负荷。为了应对这一挑战,基于多样化的物联网应用需求,通过最优化设备传输的选择控制,设计了一种任务协同迁移策略,以实现时延约束下的系统最小能量消耗。在缺少信道状态完美先验信息的条件下,提出了一种基于深度增强学习的资源管理算法,以较低的复杂度获得了最优的任务卸载决策。仿真结果表明,与随机的传输选择策略相比,所提出的算法能够显著地降低系统的能量消耗,并且满足任务的服务时延。
Mobile edge computing can reduce transmission delay and data processing delay for IoT applications by executing communication and computing operation in the edge network. However, for a large number of IoT device connections, massive service data is simultaneously gathered on the edge computing platform, which will significantly increase the traffic load of the forward link and the computing load of the edge server. In order to meet this challenge, based on diversified IoT application requirements, a task collaborative migration strategy was designed to realize the minimum energy consumption of the system under time delay constraints by optimizing the selection control of equipment transmission. In the absence of perfect channel state prior information, a resource management algorithm based on deep reinforcement learning was proposed to obtain the optimal offloading decision with lower complexity. The simulation results show that the proposed algorithm can significantly reduce the energy consumption of the system and meet the service delay of the task compared with the random transmission selection strategy.
Mobile edge computing can reduce transmission delay and data processing delay for IoT applications by executing communication and computing operation in the edge network. However, for a large number of IoT device connections, massive service data is simultaneously gathered on the edge computing platform, which will significantly increase the traffic load of the forward link and the computing load of the edge server. In order to meet this challenge, based on diversified IoT application requirements, a task collaborative migration strategy was designed to realize the minimum energy consumption of the system under time delay constraints by optimizing the selection control of equipment transmission. In the absence of perfect channel state prior information, a resource management algorithm based on deep reinforcement learning was proposed to obtain the optimal offloading decision with lower complexity. The simulation results show that the proposed algorithm can significantly reduce the energy consumption of the system and meet the service delay of the task compared with the random transmission selection strategy.
引文
[1]GU Y,ZHENG C,MIAO P,et al.Joint radio and computational resource allocation in IoT fog computing[J].IEEE Transactions on Vehicular Technology,2018(99):1.
[2]SINGH A,VINIOTIS Y.An SLA-based resource allocation for IoTapplications in cloud environments[C]//Cloudification of the Internet of Things.2017.
[3]ABBAS N,ZHANG Y,TAHERKORDI A,et al.Mobile edge computing:a survey[J].IEEE Internet of Things Journal,2017,(99):1.
[4]BELLAVISTA P,CHESSA S,FOSCHINI L,et al.Human-enabled edge computing:exploiting the crowd as a dynamic extension of mobile edge computing[J].IEEE Communications Magazine,2018,56(1):145-155.
[5]ZHAO T,ZHOU S,GUO X,et al.Tasks scheduling and resource allocation in heterogeneous cloud for delay-bounded mobile edge computing[C]//IEEE International Conference on Communications.IEEE,2017.
[6]MAO Y Y,ZHANG J,SONG S H,et al.Power-delay tradeoff in multi-user mobile-edge computing systems[C]//2016 IEEE Global Communications Conference.IEEE,2016.
[7]YU Y H,ZHANG J,LETAIEF K B.Joint subcarrier and CPU time allocation for mobile edge computing[C]//2016 IEEE Global Communications Conference(GLOBECOM).IEEE,2016:1-6.
[8]WANG C,YU F R,CHEN Q,et al.Joint computation and radio resource management for cellular networks with mobile edge com-puting[C]//IEEE International Conference on Communications.IEEE,2017.
[9]WANG Y,MIN S,WANG X,et al.Mobile-edge computing:partial computation offloading using dynamic voltage scaling[J].IEEETransactions on Communications,2016,64(10):4268-4282.
[10]XU J,REN S.Online learning for offloading and autoscaling in renewable-powered mobile edge computing[C]//2016 IEEE Global Communications Conference(GLOBECOM).IEEE,2017.
[11]WANG C,LIANG C,YU F R,et al.Computation offloading and resource allocation in wireless cellular networks with mobile edge computing[J].IEEE Transactions on Wireless Communications,2017,16(8):4924-4938.
[12]CHATTERJEE P,GHOSH S C,DAS N.Load balanced coverage with graded node deployment in wireless sensor networks[J].IEEE Transactions on Multi-Scale Computing Systems,2017,(99):1.
[13]MENDEL J M.Fuzzy logic systems for engineering:a tutorial[J].Proc of the IEEE,1995,83(3):345-377.
[14]LEVIN E,PIERACCINI R,ECKERT W.Using Markov decision process for learning dialogue strategies[C]//IEEE International Conference on Acoustics.IEEE,1998.
[15]JEONG S,SIMEONE O,KANG J.Mobile edge computing via a UAV-mounted cloudlet:optimization of bit allocation and path planning[J].IEEE Transactions on Vehicular Technology,2018,67(3):2049-2063.
[16]XU C,PU L,LIN G,et al.Exploiting massive D2D collaboration for energy-efficient mobile edge computing[J].IEEE Wireless Communications,2017,24(4):64-71.
[17]HE L,OTA K,DONG M.Learning IoT in edge:deep learning for the Internet of things with edge computing[J].IEEE Network,2018,32(1):96-101.
[18]LIN C T.A neural fuzzy control system with structure and parameter learning[J].Part I Fuzzy Sets&Systems,1995,70(2):183-212.
[19]CAO S G,REES N W,FENG G.Analysis and design of uncertain fuzzy control systems.Part I fuzzy modelling and identification[C]//IEEE International Conference on Fuzzy Systems.1996.
[20]RIEDMILLER M.Neural fitted Q iteration-first experiences with a data efficient neural reinforcement learning method[C]//European Conference on Machine Learning.2005.
[21]BOTVINICK M M,NIV Y,BARTO A C.Hierarchically organized behavior and its neural foundations:a reinforcement learning perspective[J].Cognition,2009,113(3):262-280.
[2]SINGH A,VINIOTIS Y.An SLA-based resource allocation for IoTapplications in cloud environments[C]//Cloudification of the Internet of Things.2017.
[3]ABBAS N,ZHANG Y,TAHERKORDI A,et al.Mobile edge computing:a survey[J].IEEE Internet of Things Journal,2017,(99):1.
[4]BELLAVISTA P,CHESSA S,FOSCHINI L,et al.Human-enabled edge computing:exploiting the crowd as a dynamic extension of mobile edge computing[J].IEEE Communications Magazine,2018,56(1):145-155.
[5]ZHAO T,ZHOU S,GUO X,et al.Tasks scheduling and resource allocation in heterogeneous cloud for delay-bounded mobile edge computing[C]//IEEE International Conference on Communications.IEEE,2017.
[6]MAO Y Y,ZHANG J,SONG S H,et al.Power-delay tradeoff in multi-user mobile-edge computing systems[C]//2016 IEEE Global Communications Conference.IEEE,2016.
[7]YU Y H,ZHANG J,LETAIEF K B.Joint subcarrier and CPU time allocation for mobile edge computing[C]//2016 IEEE Global Communications Conference(GLOBECOM).IEEE,2016:1-6.
[8]WANG C,YU F R,CHEN Q,et al.Joint computation and radio resource management for cellular networks with mobile edge com-puting[C]//IEEE International Conference on Communications.IEEE,2017.
[9]WANG Y,MIN S,WANG X,et al.Mobile-edge computing:partial computation offloading using dynamic voltage scaling[J].IEEETransactions on Communications,2016,64(10):4268-4282.
[10]XU J,REN S.Online learning for offloading and autoscaling in renewable-powered mobile edge computing[C]//2016 IEEE Global Communications Conference(GLOBECOM).IEEE,2017.
[11]WANG C,LIANG C,YU F R,et al.Computation offloading and resource allocation in wireless cellular networks with mobile edge computing[J].IEEE Transactions on Wireless Communications,2017,16(8):4924-4938.
[12]CHATTERJEE P,GHOSH S C,DAS N.Load balanced coverage with graded node deployment in wireless sensor networks[J].IEEE Transactions on Multi-Scale Computing Systems,2017,(99):1.
[13]MENDEL J M.Fuzzy logic systems for engineering:a tutorial[J].Proc of the IEEE,1995,83(3):345-377.
[14]LEVIN E,PIERACCINI R,ECKERT W.Using Markov decision process for learning dialogue strategies[C]//IEEE International Conference on Acoustics.IEEE,1998.
[15]JEONG S,SIMEONE O,KANG J.Mobile edge computing via a UAV-mounted cloudlet:optimization of bit allocation and path planning[J].IEEE Transactions on Vehicular Technology,2018,67(3):2049-2063.
[16]XU C,PU L,LIN G,et al.Exploiting massive D2D collaboration for energy-efficient mobile edge computing[J].IEEE Wireless Communications,2017,24(4):64-71.
[17]HE L,OTA K,DONG M.Learning IoT in edge:deep learning for the Internet of things with edge computing[J].IEEE Network,2018,32(1):96-101.
[18]LIN C T.A neural fuzzy control system with structure and parameter learning[J].Part I Fuzzy Sets&Systems,1995,70(2):183-212.
[19]CAO S G,REES N W,FENG G.Analysis and design of uncertain fuzzy control systems.Part I fuzzy modelling and identification[C]//IEEE International Conference on Fuzzy Systems.1996.
[20]RIEDMILLER M.Neural fitted Q iteration-first experiences with a data efficient neural reinforcement learning method[C]//European Conference on Machine Learning.2005.
[21]BOTVINICK M M,NIV Y,BARTO A C.Hierarchically organized behavior and its neural foundations:a reinforcement learning perspective[J].Cognition,2009,113(3):262-280.