Prediction of Time Delay in eLoran Data Acquisition Systems

This study examines the influence of six meteorological factors on eLoran signal propagation delay and develops a predictive model using four machine learning approaches: BP neural networks, random forests, support vector regression, and gradient boosting. Optimization of these models led to significant reductions in root mean square error (RMSE) and mean absolute error (MAE). Results demonstrate that the gradient boosting model offers superior accuracy in predicting propagation delay, satisfying the eLoran system's timing requirement of 100 ns. This research provides valuable theoretical insights and practical guidance for enhancing the eLoran system's application in interference-resistant environments.