| Time Sensitive Networking(TSN)is a new technology that can provide deterministic and reliable transmission for related real-time systems.TSN allows mixed communication of Time-triggered(TT)streams and non-TT streams(Audio Video Bridging(AVB)streams and Best Effort(BE)streams),and can provide globally scheduled transmission messages to provide bounded delays for time-triggered communications,this saves costs and preserves backward compatibility.In the case of mixed multi-dimensional data streams,most of the current TSN scheduling algorithms provide solutions from the perspective of scheduling constraints,and do not consider the impact of traffic transmission paths on the scheduling problem.With the continuous expansion of the application scale of TSN,there are more and more types of data streams and optional paths transmitted in the network.TSN urgently needs to solve the problem of path planning and scheduling in the mixed transmission of TT streams and non-TT streams.Therefore,this thesis proposes a joint optimization method of TSN path selection and scheduling,which combines the resourceaware KPA path selection algorithm and the constraint-based gated scheduling method.The main research work of this thesis is as follows:1.Firstly,this thesis analyzes the different delay requirements and transmission characteristics of TT streams and non-TT streams in the transmission process,defines the necessary factors affecting the solution of path planning,and proposes a resource-aware KPA(K-Shortest Path Algorithm,Path Criticality,Ant Colony Optimization)path selection algorithm.Considering the diversity of input path samples,this thesis first uses the KSP to select K shortest paths as candidate paths in the alternative path stage,and then quantifies the performance indicators of each path through the path criticality mechanism and arrange them in descending order,and then select the first m paths from the K alternative paths as the preferred paths,which realizes the further optimization of the path samples;finally,in the path selection stage,the improved ant colony algorithm is used to calculate the optimal TT streams from the sender to the receiver.The transmission path is also planned for the non-TT streams,thereby improving the transmission efficiency of the network.2.Aiming at the traffic conflict caused by overlapping transmission links in the process of TT streams transmission,after completing the transmission path planning of TT streams and non-TT streams,this thesis uses the gated scheduling mechanism and scheduling constraints to propose a gated scheduling method based on constraints.This method effectively solves the transmission uncertainty of TT streams in TSN multi-path transmission scenario by calculating the gated list under traffic scheduling constraints,it meets the requirements of TT streams for low delay and low jitter.3.In this thesis,a test and verification platform for the joint optimization method of TSN path selection and scheduling is built,and based on this platform,the resource-aware KPA path selection algorithm and the constraint-based gated scheduling method are verified.The verification results show that the joint optimization method of TSN path selection and scheduling reduces the delay of the TT streams by 44% and the jitter by 87%,ensuring the deterministic type of the critical data stream in TSN.Finally,the joint optimization method of TSN path selection and scheduling is compared and verified through the running time and schedulability of the algorithm. |
