Research On Offloading Method For Optimizing Energy Consumption And Delay In Mobile Edge Computing

Compared to traditional Cloud Computing(CC),Mobile Edge Computing(MEC)pays more attention to extending computing and storage resources to the edge of network that is close to users,which not only alleviates the shortage of hardware resources for mobile devices,but also avoids the heavy load and high delay caused by the transmission of a large amount of data in the core network.However,this new computing paradigm also brings some new challenges,among which computational offloading is considered to be one of the most important challenges.This paper investigates the research actuality of computational offloading in MEC system,and summarizes two typical MEC computing offloading scenarios: user-oriented MEC computational offloading scenarios and operator-oriented MEC computational offloading scenarios.The former focuses on improving the quality of service(Qo S)of MEC service subscribers,while the latter pays more attention to providing better services for more service subscribers at less cost.In this paper,we will study computational offloading policy in two typical MEC scenarios for optimizing latency and energy consumption.The main work of this paper is as follows:(1)In the user-oriented MEC computational offloading scenario,for the complex dependencies among tasks of some new mobile applications,a layered computational offloading algorithm based on topological sorting is proposed to layer tasks so that there is no dependency among tasks in the same layer.In order to get the computational offloading policy,the computing costs of the task on local device and MEC server,which are related to resource allocation scheme,are defined,respectively.The offloading decision and resource allocation scheme of each task are obtained by obtaining and comparing the minimum computing cost of tasks on the local device and MEC servers.Finally,simulation experiments are designed to verify that the proposed algorithm outperforms other comparison algorithms in optimizing application delay and the mobile device energy consumption.(2)In the operator-oriented MEC computational offloading scenario,aiming at the problem of large solution space of offloading decision and resource competition and mutual interference among tasks in large-scale MEC system,this paper considers dividing the original problem into two sub-problems to get the near-optimal computational offloading policy.For the resource allocation sub-problem,the closed-form solution of the optimal resource allocation scheme under any given task offloading decision is given,so the original problem is transformed into getting the optimal task offloading decision.In order to solve the problem that the solution space of the task placement sub-problem is too large,the Ordinal Optimization(OO)theory is introduced,and the near-optimal task offloading policy is obtained with high probability.Finally,simulation experiments are designed to verify that the proposed algorithm outperforms other comparison algorithms in optimizing task delay and MEC server energy consumption.