ELoran Propagation Delay Prediction Model Based on a BP Neural Network for a Complex Meteorological Environment

Area of Science:

• Navigation Systems
• Geophysics
• Artificial Intelligence

Background:

• eLoran ground-based timing navigation systems rely on accurate groundwave propagation delay measurements.
• Meteorological changes significantly impact propagation delay, especially in complex terrestrial environments, affecting system timing accuracy.

Purpose of the Study:

• To develop a robust propagation delay prediction model for complex meteorological conditions.
• To mitigate the impact of weather-induced timing inaccuracies in eLoran systems.

Main Methods:

• Theoretical analysis of meteorological factors influencing propagation delay components.
• Correlation analysis of measured data to identify relationships between meteorological factors and propagation delay.
• Development and validation of a Back-Propagation Neural Network (BPNN) model incorporating regional meteorological variations.

Main Results:

• Demonstrated complex relationships between seven key meteorological factors and propagation delay, including regional differences.
• The proposed BPNN model effectively predicts propagation delay fluctuations over several days.
• Significant performance improvement compared to existing linear and simple neural network models.

Conclusions:

• The BPNN model offers a viable solution for predicting and compensating for meteorological disturbances in eLoran timing systems.
• Accurate prediction of propagation delay fluctuations enhances the overall reliability and precision of ground-based timing navigation.