Research On Joint Optimization Of Task Offloading In Mobile Edge Computing

With the rapid development of mobile communication technology and Internet technology,the number of network user terminal devices continues to increase,and applications have put forward higher requirements in terms of performance indicators such as delay,energy consumption and throughput.Terminal equipment cannot meet the above requirements due to limited battery life and processing capabilities,so mobile edge computing needs to be considered.Under the mobile edge computing architecture,the terminal can offload tasks to the edge layer or the cloud for computing,achieving the goals of lower latency,lower energy consumption and higher throughput.Therefore,how to properly perform task offloading and resource allocation is an important issue that needs to be solved urgently in mobile edge computing.The main work of the thesis is as follows:Firstly,the joint optimization problem of energy consumption and computational throughput is proposed,and a two-timescale algorithm is used to solve it effectively.In mobile edge computing,the vertical cooperation between edge servers and cloud servers and the horizontal cooperation between edge servers are considered at the same time,a communication model and server task queue execution model are constructed,and a joint optimization problem of energy consumption and computing throughput is proposed which is a mixed integer nonlinear problem.Consider using a twotimescale algorithm framework to solve this problem,and predict the flow from the terminal side to the edge side through the long and short-term memory network on a large time scale,which is used to update the task queue of the edge server;on a small time scale,the Lyapunov method is used to transform the joint optimization problem into a steady optimization problem,and then the problem is decomposed into multiple sub-problems: communication resource optimization allocation problem,server computing resource optimization allocation problem and scheduling optimization problem between edge servers.Finally,the optimal resource allocation strategy and scheduling strategy of the joint optimization problem are solved.The simulation results show that the proposed scheme can significantly reduce energy costs and increase task processing speed.Secondly,the joint optimization problem of delay and load balancing is proposed,and the fireworks algorithm is used to solve it effectively.Considering the heterogeneity of computing tasks and edge servers,a mobile edge computing communication model and computing model are constructed,and a load balancing and delay joint optimization problem is proposed.In problem solving,tasks are clustered and analyzed according to their characteristics such as bandwidth and storage to achieve cluster division.Using the new selection strategy in the fireworks algorithm and the firework explosion radius detection mechanism,the terminals offload task clusters with high resource requirements to the cloud server for processing,and offload task clusters with relatively small resource requirements to edge servers for processing.The simulation results show that the proposed scheme can effectively achieve load balancing and reduce the task completion delay.