Research On Time-triggered Traffic Scheduling For Time-Sensitive Networks In Industrial Internet Of Things

With the continuous development of Industrial information technology in China,the Industrial Internet of Things(IIo T)is also showing a booming trend.Various sensors,mobile communication and intelligent analysis technologies are constantly integrated into industrial production,resulting in the rapid increase of applications and services carried by them.A large number of data flows pour into applications,resulting in a rapid increase in the size of data flows.Especially for some tasks in the field of industrial control,the transmission of data flows requires strict real-time communication requirements and has subtle jitter and delay.However,the complex special field bus and traditional ethernet are difficult to meet the complex industrial environment.Under the industrial internet of things scenario,a real-time communication network is urgently needed for data transmission.Time-Sensitive Networking(TSN),as a new industrial communication technology,meets the needs of industrial scenarios and is gradually being studied and adopted by the industry.The core mechanism of TSN communication is mainly traffic scheduling.In the scenario where large-scale industrial data needs to be transmitted,reasonable scheduling are required to effectively avoid conflicts and adverse consequences.However,most of the current research only considers static scheduling in small-scale scenarios,and it is difficult to achieve scaling in large-scale dynamic application scenarios.In addition,most of the existing scheduling methods only focus on scheduling,ignoring joint routing and scheduling.The traditional integer linear programming method solves the scheduling speed slowly.And unknown network link failures also bring certain difficulties to the real-time communication transmission of the network.This paper focuses on the scheduling problem of time-triggered(TT)flow for time-sensitive networks in the IIo T scenario.The main research content and innovation points are as follows:(1)Aiming at the problem that the existing scheduling methods are difficult to carry out scalable dynamic real-time TT flow scheduling in IIo T scenarios,this paper proposes a scalable scheduling method based on dynamic online grouping in industrial time-sensitive networks.In this method,this paper innovatively proposes a dynamic online grouping method.Based on the conflict index between TT flows.We establish an undirected weighted flow graph,divide the time into equally spaced time windows,and dynamically group TT flows online locally within each time window.This makes the flow scheduling problem scalable to solve,when the flow to be scheduled to multiply,can achieve scalable and efficient solution.This scheduling method can efficiently schedule dynamically arriving TT flows in industrial scenarios,avoid unnecessary conflicts in the data communication process,and achieve low-latency transmission.(2)The current research only considers the scheduling problem and does not consider joint routing scheduling problem and the impact of the unknown network link fault on the real-time communication.This paper proposes a meta-heuristic based multipath joint routing and scheduling method.This method jointly plans the routing and scheduling,considers the redundant path transmission,and establishes an integer linear programming model to separate time and space.Finally,the meta-heuristic algorithm is used to obtain a high-quality solution and achieve reliable real-time communication of TT flows.(3)The proposed a scalable scheduling method based on dynamic online grouping in industrial time-sensitive networks and a meta-heuristic based multipath joint routing and scheduling method are tested in this paper.The efficiency of the two scheduling methods proposed in this paper is verified through experiments and performance analysis.Experiments show that the first method proposed in this paper can significantly increase the number of scheduling flows and obtain high quality scheduling solutions under scalable large-scale dynamic data flow conditions,and the runtime is significantly reduced.The second method proposed in this paper can obtain high-quality solutions with low runtime,thereby achieving low-latency transmission communication of real-time data in industrial scenarios.In this paper,a scalable scheduling method based on dynamic online grouping in industrial time-sensitive networks and a meta-heuristic based multipath joint routing and scheduling method are used to efficiently schedule real-time application data in industrial scenarios,which significantly increase the number of TT flows in the network to obtain high quality scheduling solutions and effectively reduce the runtime.