Mobile Edge Computing-based Task Co-offloading Delay And Energy Performance Study

With the advent of the mobile Internet era,the demand of user for data processing speed and Quality of Service(Qo S)is growing exponentially.Despite the increasing performance of central processors in mobile terminals,their computing power is still insufficient to cope with the rapid growth of tasks,which has led to the emergence and development of Mobile Edge Computing(MEC).The MEC provides better services to users by deploying edge devices at the network edge,allowing users to offload computationally intensive tasks,reducing the time and energy required for users’ computation tasks.Based on this,this paper investigates the delay and energy consumption performance of MEC-based collaborative task offloading.The main contributions of this thesis are as follows:Firstly,a MEC system consisting of a single MEC server for multiple edge devices is studied,which includes a cellular base station configured with a MEC server,multiple edge devices,and a single user,where the MEC server and the edge devices are jointly used as a data processing center and the user is considered not to perform local computation.For the optimization problem of the total processing delay of this system,a joint allocation scheme of the MEC-based task allocation ratio and user transmit power is proposed to solve the optimization problem with the objective of minimizing the total processing delay of the system.Simulation results show that the proposed scheme reduces the system processing delay compared to the equal task allocation scheme,equal power allocation scheme,and no edge device assisted offloading scheme.Then,a MEC system based on multiple edge devices and single users,which includes multiple edge devices and a single user,where the edge devices and users jointly serve as data processing centers.For the optimization problem of total system energy consumption,a joint allocation scheme of task allocation ratio,user transmit power,and bandwidth based on MEC is proposed.This problem is solved by an alternating optimization algorithm with the objective of minimizing total system energy consumption.Simulation results show that the proposed scheme significantly reduces the total energy consumption of the system compared with the equal task,equal power,and equal bandwidth allocation scheme.Finally,a single user MEC system with multiple edge devices assisted by Reconfigurable Intelligent Surface(RIS)is studied,which includes multiple edge devices,a single user and a single RIS node,the edge devices and the user act as the data processing center together.For the optimization of the system offload data volume,a RIS-assisted user task offload maximization method is proposed,and an iterative algorithm based on alternating optimization of the receiving beamforming vector,offload time slot assignment,RIS phase shift,and user transmit power is designed to improve the total system offload data volume.Simulation results show that the proposed scheme significantly increases the amount of user task offload data compared to the conventional no RIS-assisted task offload scheme.