Research On Scheduling Model Of Edge Computing Tasks Based On Users Social Relations

With the increasing number of mobile devices and the emergence of mobile applications with high demand for computing resources,traditional cloud computing cannot provide low-latency and high-quality computing services for mobile devices with limited resources.Edge computing emerges as the times require.It schedules the tasks of mobile devices to the edge server close to the user,reduces the response time,relives the pressure of network bandwidth,and brings better experience for users.Using mobile devices as edge servers,through quantifying the social relationships among device users,trusted mobile servers can be found for task offloading,ensuring safe and reliable task unloading and preventing privacy leakage.D2D communication provides good technical support for task unloading,and energy saving task unloading can be realized by small transmitting power between mobile devices.On the basis of summarizing and analyzing the problem of tasks scheduling in edge computing,this thesis aims to reduce the time delay and energy consumption of tasks execution in the whole scene,and comprehensively considers the user’s social relations and D2D communication conditions,and puts forward the framework of task scheduling model for edge computing based on social relations.Firstly,the social relations among users are analyzed and quantified,and the social relations graph of users is constructed based on mathematical convolution and Jaccard similarity coefficient.In order to ensure that social relationships are accurately quantified to provide safe and reliable task scheduling,the graph describes users’social relationships using mathematical convolution and Jaccard similarity coefficient.Then,the users’device connection graph is constructed based on Euclidean distance to ensure that the tasks scheduling meets the D2D communication conditions.Next,the users’tasks-service devices bipartite graph is constructed by combining the above two graphs,and the task execution cost is calculated to determine the weight of the edge of the bipartite graph,thus completing the construction of the task scheduling model.Finally,the framework of tasks scheduling model in edge computing based on social relations is implemented,and the tasks scheduling model in edge computing is solved based on KM algorithm.The experimental results show that the model and algorithm proposed in this thesis have a good effect on solving the problem of tasks scheduling in edge computing,which can reduce the execution delay and energy consumption of the whole scene,and ensure the security and reliability of tasks scheduling.