Research On Reliable Routing And Scheduling In Time-sensitive Networks

With the rise of Internet of Things and Cyber-Physical System,great changes are taking place in the field of industrial communication.Real-time Ethernet technology based on the evolution of general Ethernet technology has gradually replaced the traditional industrial bus technology and become a research hotspot.Nowadays,real-time Ethernet has become the de facto standard in the field of industrial communication,among which Time Sensitive Network(TSN)has a good prospect.TSN is a real-time communication solution that abstracts transmission tasks with deterministic delay and jitter into time-sensitive data flows and provides a deterministic quality of service(Qo S)for switched Ethernet.Routing and scheduling for time-sensitive data flows in TSN with limited resources is a hot research problem.Existing work only considers scheduling on a single transmission path.However,in the case of ultra-reliable scenarios such as autonomous driving,it is necessary to establish redundant transmission paths as well as schedule on these paths to ensure the reliability of data transmission.Redundant transmission paths lead to heavy network load and make scheduling more difficult.In this paper,we consider the problem of reliable routing and scheduling of time-sensitive data flows in resource constrained TSNs.The main researches in this paper are as follows:(1)We define the reliable routing and scheduling problem in TSN and adopt path redundancy to ensure reliability.The goal of the problem is to maximize routing and scheduling success rate and minimize bottleneck link load while ensuring reliable delivery with bound delays and jitters.(2)For the reliable routing problem of time-sensitive data flows,we propose Redundant Paths Search(RPS)algorithm.For the scheduling problem of time-sensitive data flows with redundant routes,we propose Redundant Flows Scheduling(RFS)algorithm.Based on RPS and RFS,we propose Joint Redundant Routing and Scheduling Optimization(JRRSO)algorithm to optimize the feasible solutions generated in the routing and scheduling phases.Through experiments on open source TSN network simulator and self-developed TSN testbed,it is proved that the proposed algorithm can improve the reliability of data transmission while ensuring the deterministic quality of service.(3)Referring to the TSN standard suite,we design and implement the TSN testbed based on Linux operation system.The testbed realizes the core functions of TSN standard,and improves the scheduling mechanism of TSN standard to make the scheduling adapt to random dynamic events in the actual network.The testbed can be deployed to real physical machine for experimental testing of routing and scheduling algorithms.