A Railway Train Number Tracking Method Using a Prediction Approach

Obtaining the accurate real-time train number and location in a railway network is necessary for railway traffic operation control. This paper investigates the implementation and optimization method of train number tracking, by which the train number and location information can be captured. According to the characteristics of the train operation, the train moving trajectory and the train number tracking problem, the mathematical description of the problem and the tracking model based on the railway signaling states and train schedules are proposed. Then, a method using a hidden Markov model prediction is proposed in order to improve the correctness of train number tracking. The simulation results are compared with the results obtained under certain restrictions, and the analyses are discussed. The results show that the proposed method can effectively improve the accuracy of train number tracking with better fault-tolerant robustness.