The combination of energy harvesting small cell networks (EH-SCNs) and mobile edge computing (MEC) has been considered as an effective means to improve the performance of mobile networks and provide users with a highe...
详细信息
The combination of energy harvesting small cell networks (EH-SCNs) and mobile edge computing (MEC) has been considered as an effective means to improve the performance of mobile networks and provide users with a higher quality of service (QoS). In this letter, we investigate the decentralized computationoffloading problem in heterogeneous EH-SCNs with MEC, where heterogeneous small cell base stations (SBSs) are rational individuals with interests to maximize their own benefits while considering their QoS requirements. Different from existing works, we address the challenge that heterogeneous SBSs may unwilling to expose their own information about the system state and offloading decisions. We formulate the problem as a partially observable stochastic game (POSG), in which SBSs can make optimal offloading decisions with imperfect state information. We analyze the local equilibrium, and propose a stochastic offloading algorithm to obtain the approximate optimal solution. Numerical results validate the effectiveness of the proposed scheme.
To meet the demands of large-scale user access with computation-intensive and delay-sensitive applications,combining ultra-dense networks(UDNs)and mobile edge computing(MEC)are considered as important *** the MEC enab...
详细信息
To meet the demands of large-scale user access with computation-intensive and delay-sensitive applications,combining ultra-dense networks(UDNs)and mobile edge computing(MEC)are considered as important *** the MEC enabled UDNs,one of the most important issues is computation *** a number of work have been done toward this issue,the problem of dynamic computation offloading in time-varying environment,especially the dynamic computation offloading problem for multi-user,has not been fully ***,in order to fill this gap,the dynamic computation offloading problem in time-varying environment for multi-user is considered in this *** considering the dynamic changes of channel state and users’queue state,the dynamic computation offloading problem for multi-user is formulated as a stochastic game,which aims to optimize the delay and packet loss rate of *** find the optimal solution of the formulated optimization problem,Nash Q-learning(NQLN)algorithm is proposed which can be quickly converged to a Nash equilibrium ***,extensive simulation results are presented to demonstrate the superiority of NQLN *** is shown that NQLN algorithm has better optimization performance than the benchmark schemes.
With the rapid development of Mobile Edge Computing (MEC) technology, the computationally intensive requests of end devices can be offloaded to MEC servers directly, which equipped at the edge of wireless networks. Th...
详细信息
With the rapid development of Mobile Edge Computing (MEC) technology, the computationally intensive requests of end devices can be offloaded to MEC servers directly, which equipped at the edge of wireless networks. Through offloading, the performances such as the execution delay as well as the energy consumption can be effectively improved, which can significantly enhance the quality of user experience. Given the dynamics and randomness of computation requests arrival, the energy in the battery, the radio network environment, and the computation resource in the MEC server, it is a challenge to perform efficient offloading. Based on these problems, this article proposes dynamic optimization schemes with queuing theory for the cases of the static subchannel and dynamic subchannel during a time slot separately in 5G MEC heterogeneous networks with multiple MDs equipped with the function of energy harvesting. In the schemes, offloading decisions and radio allocation strategies will be dynamically coordinated. They are also jointly allocated along with changing wireless communication resources and computation demands aiming to minimize the system average execution delay. Specifically, it is assumed that the offloading requests can be transmitted through either macro base stations or small base stations. In the case of the static subchannel, a joint resource allocation and computationoffloading scheme based on Lyapunov optimization and Simulated Annealing Genetic Algorithm (SAGA) is put forward. As for the dynamic subchannel, the leader-and-follower model is adopted and solved by SAGA and Sequential Quadratic Programming (SQP) method. At last, the effectiveness of the proposed schemes is verified through several simulations.
Driven by the tremendous in vehicular networks computation-intensive application demands, the incorporation of mobile edge computing (MEC) and vehicular cloud is convinced as a promising paradigm to fulfill computatio...
详细信息
ISBN:
(纸本)9781728189642
Driven by the tremendous in vehicular networks computation-intensive application demands, the incorporation of mobile edge computing (MEC) and vehicular cloud is convinced as a promising paradigm to fulfill computationoffloading requirements. However, the changing vehicular communication topology (CVCT) poses a significant challenge for offloading directed acyclic graph (DAG) model application. Due to the precedence and connection constraint between different sub-jobs, the successful offloading of DAG-enabled apllication will be disturbed even interrupted without considering CVCT. To address this problem, we propose a topology-aware dynamic computaion offloading mechanism and adopt simulated annealing algorithm (TASA) to jointly optimize the energy consumption and completion time under dynamic environment, while guaranteeing the convergence of the proposed method. Simulation results reveal the effectiveness of the proposed method in overcoming CVCT's influence.
Fog computing system is able to facilitate computation-intensive applications and emerges as one of the promising technology for realizing the Internet of Things (IoT). By offloading the computational tasks to the fog...
详细信息
Fog computing system is able to facilitate computation-intensive applications and emerges as one of the promising technology for realizing the Internet of Things (IoT). By offloading the computational tasks to the fog node (FN) at the network edge, both the service latency and energy consumption can be improved, which is significant for industrial IoT applications. However, the dynamics of computational resource usages in the FN, the radio environment and the energy in the battery of IoT devices make the offloading mechanism design become challenging. Therefore, in this article, we propose a dynamic optimization scheme for the IoT fog computing system with multiple mobile devices (MDs), where the radio and computational resources, and offloading decisions, can be dynamically coordinated and allocated with the variation of radio resources and computation demands. Specifically, with the objective to minimize the system cost related to latency, energy consumption, and weights of MDs, we propose a joint computationoffloading and radio resource allocation algorithm based on Lyapunov optimization. Through minimizing the derived upper bound of the Lyapunov drift-plus-penalty function, we divide the main problem into several subproblems at each time slot and address them accordingly. Through performance evaluation, the effectiveness of the proposed scheme can be verified.
In this work, we propose a dynamic optimization scheme for an edge computing system with multiple users, where the radio and computational resources, and offloading decisions, can be dynamically allocated with the var...
详细信息
In this work, we propose a dynamic optimization scheme for an edge computing system with multiple users, where the radio and computational resources, and offloading decisions, can be dynamically allocated with the variation of computation demands, radio channels and the computation resources. Specifically, with the objective to minimize the energy consumption of the considered system, we propose a joint computationoffloading, radio and computational resource allocation algorithm based on Lyapunov optimization. Through minimizing the derived upper bound of the Lyapunov drift-plus-penalty function, the main problem is divided into several sub-problems at each time slot and are addressed separately. The simulation results demonstrate the effectiveness of the proposed scheme.
This paper studies the computational offloading of CNN inference in dynamic multi-access edge computing (MEC) networks. To address the uncertainties in communication time and Edge servers' available capacity, we u...
详细信息
ISBN:
(纸本)9781665435406
This paper studies the computational offloading of CNN inference in dynamic multi-access edge computing (MEC) networks. To address the uncertainties in communication time and Edge servers' available capacity, we use early-exit mechanism to terminate the computation earlier to meet the deadline of inference tasks. We design a reward function to trade off the communication, computation and inference accuracy, and formulate the offloading problem of CNN inference as a maximization problem with the goal of maximizing the average inference accuracy and throughput in long term. To solve the maximization problem, we propose a graph reinforcement learningbased early-exit mechanism (GRLE), which outperforms the state-of-the-art work, deep reinforcement learning-based online offloading (DROO) and its enhanced method, DROO with early-exit mechanism (DROOE), under different dynamic scenarios. The experimental results show that GRLE achieves the average accuracy up to 3.41x over graph reinforcement learning (GRL) and 1.45x over DROOE, which shows the advantages of GRLE for offloading decision-making in dynamic MEC.
暂无评论