GNSS Signal Interference Classified by Means of a Supervised Learning Method Applied in the Time-Frequency Domain

The Global Navigation Satellite System (GNSS) signals received by an occulting Low Earth Orbiter (LEO) often reflect on the Earth’s surface, thus interfering with the direct ray. For certain remote sensing applications, the identification and classification of the reflection events is required. We have taken advantage of the fact that interference patterns in the time-domain (oscillation of both received amplitude and phase) yield a clear feature in the frequency-time domain, to apply pattern recognition techniques onto the radioholographic images of the received electromagnetic field.