Research On Task Offloading For Low-Power In MEC

In recent years,with the rapid growth of the number of mobile smart devices,the massive amount of data that users used to process and the computationally intensive processing tasks have placed higher demands and challenges on traditional network and computing architectures.In the traditional cloud computing mode,the processing task is generally offloaded to the cloud data center for processing to reduce the processing delay of the task and reduce the power consumption of the local terminal.Because the cloud data center is generally deployed in the core network away from the user terminal,the massive amount of services and data are offloaded to the cloud service,which often results in high transmission delay due to network congestion.Therefore,the idea of lowering the function of the cloud platform to the edge of the network close to the user in order to reduce the processing and transmission delay of the task has gradually attracted people's attention.According to this,the researcher proposes the concept of edge computing to provide computing and processing services to the user.Mobile devices generally offload computationally intensive tasks to computingrich edge servers or collaboratively processed by other mobile devices in order to reduce task processing delays and save local energy consumption due to limited energy and computing resources..In edge calculation,the way to uninstall the task includes offloading to the MEC server and D2 D for task offloading.This paper firstly studies the dynamic task offloading problem based on MEC server in mobile edge environment.Firstly,a directed acyclic graph model with related dependencies is established for the associated tasks generated by the application segmentation.Then establish the task execution time and energy consumption model,and formally describe the low-power offload optimization problem.Aiming at the optimization problem,this paper proposes a dynamic offloading strategy based on greedy thought and a server-side task scheduling strategy based on greedy thoughts for the changing edge environment.Finally,through the simulation experiments,the results show that the strategy has s certain effect on task offloading.Then,this paper studies the problem of offloading based on D2 D.D2D task offloading uses D2 D communication technology to detect nearby available idle mobile devices,by offloading tasks to the available mobile devices it detects to reduce task processing latency and local power consumption.First,a communication topology diagram composed of a user equipment,a relay forwarding device,and a service device is established through D2 D communication technology.Then,based on the task attributes,communication delays,etc.,the offloading environment is estimated,and the auction model is used to select the location where the task is offloading.Finally,through simulation experiments,the availability of this strategy for D2 D task offloading is provided,which provides a new idea for D2 D task offloading.