AI for Ionospheric Disturbance Analysis Using GNSS TEC Data

Abstract

The ionosphere, a dynamic layer of Earth’s atmosphere, significantly impacts radio wave propagation and GNSS signal integrity, influencing communication, navigation, and space weather systems. This review explores recent advancements in artificial intelligence (AI) for analyzing ionospheric disturbances using Global Navigation Satellite System (GNSS) Total Electron Content (TEC) data. Machine learning algorithms—including supervised, unsupervised, and reinforcement learning—are highlighted for their roles in pattern detection, classification, and prediction of ionospheric anomalies. Deep learning architectures, such as Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks, demonstrate exceptional capability in processing large-scale TEC datasets, enabling high-resolution ionospheric mapping and temporal forecasting. Hybrid models that integrate AI with traditional ionospheric frameworks further enhance predictive accuracy and real-time adaptability. These advancements not only deepen our understanding of ionospheric dynamics but also provide actionable tools to mitigate disruptions in critical technologies.