Research on train time synchronization algorithm based on Kalman filter

Network time protocols cannot meet the requirements of train communication nodes for time synchronization accuracy. To address these problems, time sensitive network technology is applied to train communication networks, and its support for precise time protocols can provide sub-microsecond or even nanosecond time synchronization accuracy. In order to reduce the time offset and ensure the time accuracy of the master and slave nodes of the train communication network, a Kalman filter optimization algorithm is proposed in the improved clock servo system, where a threshold value is estimated from the previous time offset in the last calibration process of the clock, and the difference between the time reference value of the master clock and the time measurement value of the slave clock is compared to obtain the optimal value. Finally, a train communication network simulation environment is built using the OMNeT++ simulation platform to verify the correctness of the algorithm.