Millimeter Wave Systems Field Strength Prediction for High-Speed Maglev Train Based on STAR

The field strength of high-speed maglev millimeter wave (MMW) wireless communication systems is a direct reflection of system status and performance. In order to achieve state perception and fault detection under high-speed train operation conditions, it is necessary to accurately predict the field strength of the millimeter wave system. Firstly, an advanced field strength prediction model based on STAR module (FSPSTAR) is constructed for high-speed maglev MMW systems. The STAR module is utilized to extract feature dependencies between variables such as the current of the radio base station (RBS) demodulation module and the voltage of the frequency synthesizer module. Subsequently, linear projection is employed to obtain the temporal dependencies of the RBS demodulation module voltage variables. Finally, a multi-layer perceptron (MLP) module is used to integrate the relevant features and temporal characteristics of the variables to accomplish the prediction of the MMW system’s field strength. Validation using data collected from actual maglev train experiments shows that the root mean square error (RMSE) and the mean absolute error (MAE) of the FSPSTAR model’s prediction results are 51.9547 and 17.8825, respectively. The experimental results show that the model can achieve precise prediction of the field strengthof high-speed maglev train’s MMW systems.