Research On Task Scheduling And Resource Allocation Strategies For Mobile Edge Computing

With the continuous development of 5G communication technology and emerging Internet applications,the limited computing capacity and battery energy of mobile devices can hardly meet the computing and real-time requirements of Internet applications.Therefore,Mobile Edge Computing comes into being in order to solve the contradiction between users’ increasing network demands and users’ devices.Provides users with low latency,high energy efficiency,and high reliability network service experience.The computing offload technology dispatches computing tasks to appropriate MEC servers and allocates computing resources,reducing computing task completion delay and energy consumption on devices.Considering the limited computing resources of MECs and the limited power of user devices,how to design reasonable computing task scheduling and resource allocation strategies is an urgent problem to be solved.This paper studies the above issues and the specific research work is as follows:(1)In the multi-user and single-server edge computing scenario,considering the limited user device power and the situation that computing tasks queue up for service,the total system cost is designed according to the communication model and calculation model,and the task scheduling and resource allocation strategy based on energy perception is proposed.The simulation results show that the proposed strategy can optimize the task scheduling and resource allocation to dynamically adjust the optimal solution after reaching the preset power threshold of the user device.Meanwhile,compared with other algorithms,the proposed algorithm can effectively reduce the delay and energy consumption,and minimize the total system cost.(2)In the multi-user and multi-server edge computing scenario,multiple MEC servers are considered to compete to provide computing services for computing tasks requested by user devices,and SDN controller is introduced to make efficient decisions for task scheduling and resource allocation of user devices.According to the information of the computing resources of the MEC server and the computing resources requested by the user device,a pricing mechanism was designed and a DDPG-based MEC server selection algorithm was proposed to select the appropriate MEC server for the computing tasks of the user device.Then according to the dynamic computing offloading algorithm,the optimal offloading ratio of computing tasks for the user device is determined.Simulation results show that the proposed pricing strategy can make MEC server gain more revenue and reduce offloading cost of user equipment.Meanwhile,compared with other offloading algorithms,the proposed algorithm can effectively reduce the delay and energy consumption,with the reduction efficiency reaching 38.69%and 50.19%,respectively.