Research On Offloading Strategy Based On Mobile Edge Computing

The dramatically development of the two major networks of the mobile Internet and the Internet of Things has made future networks face the challenges of higher speeds,lower delays,and higher reliability.Various emerging applications make this challenge more realistic and urgent.In view of this,Mobile Edge Computing(MEC)has become a most likely network architecture for realizing the 5G vision,and has attracted widespread attention.MEC meets the high computational needs of resourceconstrained mobile devices by offloading compute-intensive tasks from mobile devices to nearby MEC servers.By offloading compute-intensive or delay-sensitive applications to nearby MEC servers,resource-constrained mobile devices can reduce execution delays and device energy consumption.The close deployment in the mobile network makes the MEC server closer to the mobile device so that the task offloads the network for faster transmission and lower energy consumption.In this case,the task offloading strategy plays a vital role.Therefore,the main research contents of this paper are as follows:First,a task offloading strategy is proposed for a single-user scenario.Joint optimization of delays and energy consumption for MEC systems with multiple independent tasks.The offloading strategy not only determines whether the task is to be offloaded,but also indicates the execution order of the task for the minimum overhead of the system.In order to reduce the time complexity,this paper proposes a sub-optimal algorithm based on binary particle swarm optimization and two-machine flow shop scheduling algorithm.Simulation results show that the proposed algorithm significantly reduces the delay and energy consumption.Second,a task offloading strategy was proposed for multi-user scenarios.In this paper,the multi-user task offloading under multi-channel wireless interference and MEC resource constrained environment is first studied.At the same time,it is proved that the optimal solution to the multi-user centralized offloading problem is NP-hard.When resources are limited,the MEC can be further offloaded and executed on the central cloud server.Therefore,a method of using game theory to achieve efficient computation and offloading in a distributed manner is adopted.We have transformed the issue of task offloading between users of mobile devices into a multi-user game.This game has a Nash equilibrium.Then,we design a two-phase task offloading algorithm,which can achieve Nash equilibrium and get the optimal solution for multiuser task offloading.Simulation results show that the algorithm can achieve higher reduction of task load performance when the user scale increases.Last,mobile edge computing is a rapidly growing field of research.However,despite the increasing number of research activities,this field lacks simulation tools that are compatible with requirements.Starting from the available emulators,a great deal of programming work is required to obtain simulation tools that meet actual needs.In order to reduce the obstacles,this article designed and implemented a simulation experiment tool called Edge Sim.Edge Sim builds on Cloud Sim to meet the specific needs of mobile edge computing research and supports the necessary capabilities in computing and network capabilities.To demonstrate Edge Sim's capabilities,this article presents a simulation using a simple usage scenario to demonstrate its capabilities.