Machine learning based Automatic Detection of Ionospheric Scintillation-like GNSS Oscillator Anomaly Using Dual Frequency Signals

In this paper, we propose a machine learning-based approach to automatically detect satellite oscillator anomaly using dual frequency signals. One major challenge is that both ionospheric scintillation and oscillator anomaly cause phase disturbance. The proposed radial basis function (RBF) support vector machine (SVM) classifier is capable of distinguishing the oscillator anomaly from scintillation. The results show that the proposed RBF SVM shows the best performance and outperform other classifiers. Compared to the RBF SVM with triple-frequency signals, the RBF SVM with dual-frequency signals shows suboptimal performance due to loss of frequency diversity, but still reaches a detection accuracy of 98.6%. In return, the proposed method can also detect oscillator anomaly from satellites that only broadcast dual-frequency signals (Block IIRM). The accurate detection performance suggests that the proposed method can be employed to a global satellite oscillator anomaly monitoring system and detect anomalies from both block IIRM and block IIF satellites.