Research On Joint Computation Offloading And Resource Allocation Strategy In Mobile Edge Computing

With the advent of 5G era,a large number of computing-intensive and delaysensitive applications require low latency and low energy consumption.However,mobile devices are limited in terms of computing resources,usually processing data at the expense of high latency and high device energy consumption.To satisfy the high service quality requirements of users on the network,mobile edge computing is proposed.It uses servers deployed at the edge of the network to provide users with computing resources,storage resources and IT services,which significantly improves the quality of services for users.As the key issues in mobile edge computing,computing offloading and resource allocation have great research value.With the goal of minimizing task execution delay and energy consumption,this paper mainly studies the joint computing offloading and resource allocation strategies in MEC system.The specific contents are as follows:In the multi-user single-cell MEC system,a joint computing offloading and resource allocation strategy is proposed.This paper first establishes a computing task offloading model,defines the task execution cost as the weighted sum of its delay and energy consumption,and establishes the optimization problem of minimizing system cost.This paper divides it into a resource allocation sub-problem and a computing offloading sub-problem to solve.The Lagrangian multiplier method is used to solve the resource allocation sub-problem,and a computational offloading algorithm based on the greedy algorithm is proposed to obtain the best offloading decision.Finally,the final solution of the problem is obtained through the joint optimization algorithm.Experimental results show that compared with other benchmark schemes,the system execution cost of the proposed algorithm can be reduced by up to 40%.In the multi-user multi-cell MEC system,a distributed coordinated joint computing offloading and resource allocation strategy is proposed.Considering that the computing resources of the MEC server in the local area are insufficient to meet user needs,load balancing can be achieved by transferring tasks to the MEC server in the nearby area.This paper first quantifies the delay and energy cost during task execution process,and models an optimization problem with the goal of minimizing system cost under the constraints of bandwidth resources and computing resources.Then the problem was transformed,we design a heuristic algorithm based on binary particle swarm,and a new transfer function was used to update the particle position to avoid falling into the local optimal solution.The experimental results show that the algorithm has good convergence,and can effectively reduce the total system cost of delay and energy consumption,so as to ensure the quality of user experience.