Detection of Coherent GNSS-R Measurements Using a Support Vector Machine

This paper presents a support vector machine (SVM) -based detection method to identify coherent reflections in GNSS-R measurements. The data was collected by Spire low-Earth orbit (LEO) satellites during January to April, 2019, and it contains the In-phase (I) and Quadrature (Q) correlation outputs at 50 Hz from open-loop (OL) tracking process. The coherence detection is investigated based on 1100 labelled 1-second data segments. The detection results show that the SVM-based method achieves a detection accuracy of 98.66% and outperforms several other algorithms.