Research And Simulation Of Clock Synchronization And Scheduling Algorithm In Time Sensitive Networking

With the rapid development of mobile Internet,the mobile business grows exponentially.Comparing with 4G,5G significantly improves the transmission rate of the mobile end,and the bandwidth consumption of 5G is estimated to be 1,000 times than that of 4G,thus,the mobile fronthaul network is under greater pressure on the aspect of data transmission.To reduce the deployment cost and improve the transmission performance of the fronthaul network,the industry is trying to introduce time sensitive networking(TSN)into the fronthaul,while solving the problems of delay,jitter and synchronization of the fronthaul network,to make use of the existing ethernet interface and switching equipment to simplify the deployment process and reduce costs.This paper researches the fronthaul network based on time-sensitive networking,and studies the clock synchronization technology and scheduling algorithm in TSN.Through the analysis of the characteristics of the fronthaul network traffic,the reason of jitter caused by fronthaul flow is clarified,and then a time-sensitive network scheduling scheme based on time division multiplexing is proposed.This scheme schedules the flow according to the transmission window,which provides high priority flow with transmission guarantee and eliminates the jitter caused by the convergence of flows and the nondeterministic background flow.In order to realize this scheme,this paper designed the time allocation-greedy(TA-Greedy)algorithm to work out a transmission window arrangement scheme that owns minimum jitter for time critical flows.In order to verify the effectiveness of the scheduling algorithm,a TSN simulation platform(TSN-NS)based on NS-3 simulator was designed and implemented in this paper.The simulation platform was used to test the existing TSN scheduling algorithm and the TA-Greedy algorithm in various scenarios,and we constructed the local clock to analyze the impact of the clock error on the algorithms.The simulation results show that the TA-Greedy algorithm is better than the existing scheduling scheme on reducing jitter.