Research On Offloading Strategy Of Edge Computing Tasks In Time-sensitive Networks

With the deep integration of Information Technology(IT)and Operation Technology(OT)in the field of intelligent manufacturing,Time Sensitive Netework(TSN)technology has received more and more attention in the manufacturing industry.Its features of providing deterministic and real-time data transmission based on Ethernet not only meet the different needs of data transmission in industrial manufacturing equipment,but also solve the problem of coexistence of multiple heterogeneous buses and protocols in the industrial field.At the same time,with more and more production equipment and Io T nodes connected to the industrial network,the amount of data that needs to be processed in real time on the industrial site is increasing.The traditional method of using centralized cloud computing centers for manufacturing data processing is no longer sufficient.Industrial intelligence needs,and the edge computing model that sinks computing power to the edge of the network has been widely used in the industry in recent years.Therefore,the fusion application of edge computing and time-sensitive network technology will be the focus of research in the future.Based on the application of the fusion of the two technologies,this article focuses on the computing task offloading strategy of edge computing resource scheduling in the TSN network in view of the differentiated features of the computing and storage capabilities of the edge computing equipment in the industrial field.On the basis of in-depth research,a decision-making model for offloading of edge computing tasks in TSN was constructed,and the transmission delay of offloading tasks in TSN was analyzed.Finally,a decision-making algorithm for task offloading of multiple computing nodes based on game theory was proposed,and sufficient simulation verification was carried out.The main research and innovations of this paper are as follows:1.Analyzed and defined the core attributes of edge computing task description,task transmission and variable parameters of parallel computing process,studied and proposed the task offloading delay and energy consumption constraint conditions in TSN,discussed computing resource management and scheduling methods,The mathematical model of task unloading in edge computing node is further constructed.2.Aiming at the task flow transmission in the process of computing task unloading,the transmission delay under TSN ieee802.qbv protocol is analyzed and the mathematical model is constructed,including the worst-case analysis of data flow transmission delay in TSN and the average queuing delay analysis of low priority task unloading flow based on queuing theory.3.This paper constructs a multi edge computing node task unloading decision-making model based on game theory,and proves the existence of Nash equilibrium in the game,and then designs an iterative algorithm,so that the decision-making process can achieve Nash equilibrium in multiple nodes' finite iterations.Finally,a simulation platform is built to verify the decision model.The experimental results show that the proposed model and algorithm can effectively optimize the resource allocation of edge computing in TSN,reduce the average computing delay,and provide a certain reference value for the integration of TSN and edge computing technology in intelligent manufacturing industry.