Research On Cache And Task Offload Strategy In Mobile Edge Computing Framework

In recent years,mobile data traffic has grown at an exponential rate,which poses a major challenge to future cellular networks.In addition,with the rapid development of wireless technology and the Internet of Things,more and more mobile devices,such as smart phones and wearable devices,have different wireless network access requirements for bandwidth and computing.In the future,mobile devices will become more intelligent,and applications deployed on them will require significant computing power and persistent data access.However,the development of these new applications and services is limited by the computing power,storage space,and battery life of these devices,and cannot be widely used.To solve these problems,mobile-edge computing(MEC)came into being.MEC provides users with short-latency and high-performance computing services by deploying computing nodes or servers at the edge of the network to meet users' computing needs for latency-sensitive tasks.In addition,collaborative buffering at the edge can effectively avoid congestion in the peak backhaul network and reduce the load on backhaul traffic and transmission delay.This topic intends to study the design of cache and task offload strategies based on mobile edge computing architecture.The specific research content is as follows:First,this paper proposed a physical-social-based cooperative cache scheme to motivate the data share between users.Specifically,we first define a new physical-social mapping relationship,and then the optimal caching strategy for maximizing the utility of social groups is obtained under the direct and indirect physical-social relationships.When users with direct physical-social relationships cannot meet the user's data needs,the Six Degrees of Separation is used to analyze indirect physical-social relationships between users.Then,we design an indirect social group formation algorithm to achieve optimal caching scheme under the indirect social group.Finally,simulation results show that the proposed caching scheme based on the physical-social relationship is superior to other caching schemes and can effectively meet the user's data needs.Next,in order to improve the processing efficiency of tasks under the condition of limited computation and communication resources,a joint optimization of computing offload and communication resource allocation scheme is proposed.We first proposed a collaborative computing framework for mobile tasks,where tasks can be offloaded to terminals,edge nodes,and cloud centers for processing.Then,according to the computation and communication capacities capacity of devices and edge nodes,a pipelined offload strategy is proposed to offload tasks to specific edge nodes and cloud centers.Based on the proposed offload strategy,joint the offload strategy,delivery rate and power allocation optimization problem are considered,and the total network calculation and transmission delay are minimized.Since this problem is non-convex,the classic successive convex approximation(SCA)approach is used to transform non-convex optimization problems into convex problems.Finally,simulation results show that the proposed cooperative offloading scheme is superior to other offloading schemes,and can effectively reduce the total computation and transmission delay.