Time Error Compensation with Neural Network for Time Synchronization in TSN
Haoze Yu, Guochu Shou, Xiaodong Zhang, Ledong Hu, Yaqiong Liu, Yihong Hu · 2021 7th International Conference on Computer and Communications (ICCC) · 2021
Time-sensitive networking featured with precise time synchronization, is significant in modern industrial systems for real-time guarantee. However, some industrial scenarios, such as smart grids and 5G fronthaul networks, have much higher requirements on time synchronization. This paper proposes a method to improve the synchronization performance of TSN with a neural network model. A time error compensation algorithm including timestamp data processing, neural network training, and error compensation is designed. Comparative experiment results show that this algorithm improves the stability and accuracy of time synchronization.