Research On TSN Scheduling Algorithm For Industrial Internet Of Things

The continuous advancement of network and communication technologies has had a tremendous impact on the national economy,and all countries attach great importance to the development of industrial Internet of Things.An crucial feature of Industry 4.0 is the highly time-sensitive network physical system.In order to ensure the deterministic behavior of the network physical system under control,a real-time communication network with deterministic bounded network delay and delay variation is needed to connect the sensors and execute.And controller.Therefore,how to provide deterministic real-time assurance for time-sensitive networks is an important research work.According to the analysis of the research status of time-sensitive networks at home and abroad,this paper designs a scheduling algorithm based on genetic simulated annealing and a scheduling algorithm based on ant colony system and applied them to the TSN scheduling problem.Finally,this paper also analyze some unfeasible scheduling schemes caused by the shortest path routing.The main researches are as follows:(1)Aiming at the time-triggered flow scheduling problem of time-sensitive networks in industrial Internet of Things,this paper combined with time-sensitive software-defined network(TSSDN)model,a TSN scheduling algorithm based on genetic simulated annealing algorithm(the algorithm of NGASA)is proposed.The algorithm is applied to the TSN scheduling problem to schedule and route timetriggered streams to avoid queuing of network queues.The simulation results show that the algorithm can effectively solve the TSN scheduling problem,and the convergence speed of the algorithm is obviously better than the traditional algorithm,which can provide real-time guarantee for data transmission in time-sensitive networks.(2)Considering the transmission delay of data packets in the link in the time-triggered traffic scheduling problem model of the time-sensitive network,the problem is abstracted into a binary multi-clip problem,which minimizes the transmission delay of the data packet while avoiding network queue queuing.Then combined with the idea of the minimum and maximum ant system,a TSN scheduling algorithm based on ant colony system(the algorithm of NACO)is proposed.The simulation results show that the algorithm can solve the TSN scheduling problem well,and it can avoid the algorithm falling into the local optimal solution compared with the traditional algorithm.It has better convergence speed and optimization ability,and has certain delay jitter.Changes can provide deterministic real-time assurance for time-sensitive networks.(3)In the scheduling model that considers the time-triggered flow of packet transmission delay,since each link has its capacity limitation,the partial scheduling scheme caused by the shortest path routing is not feasible,this paper checks the unschedulable scheme by introducing the Check function.Then,the algorithm of NACO execution process is adjusted in combination with the incremental strategy.A TSN scheduling algorithm with incremental strategy(the algorithm of I-NACO)is proposed and applied to the TSN scheduling problem.The simulation results show that the TSN scheduling algorithm with incremental strategy has a faster convergence speed and stability,and can adapt to large-scale scenarios well.The performance of the algorithm of I-NACO is better than the algorithm of NACO,as well as it can solve some unschedulable phenomena effectively.