Jammer Detection and Classification Using Machine Learning in GNSS Bands

Abstract

Global Navigation Satellite System (GNSS) signals are highly vulnerable to intentional jamming and unintentional radio frequency interference because of their extremely low received power at the receiver front end. Interference in GNSS frequency bands can significantly degrade positioning, navigation and timing (PNT) services used in aviation, transportation, surveying, military operations and critical infrastructure monitoring. This paper proposes a machine learning-based approach for jammer detection and classification in GNSS bands, based on time-frequency analysis of the signal and statistical feature extraction. The proposed system captures GNSS IQ samples and extracts discriminative features in the time domain and frequency domain, such as power spectral density, spectral centroid, kurtosis, spectral flatness and bandwidth occupancy. The Short Time Fourier Transform (STFT) is used to analyse the temporal and spectral characteristics of various jammer signals. Machine learning models like Support Vector Machine, Random Forest, Convolutional Neural Network, and Long Short-Term Memory networks are used to classify various types of jammers, including continuous wave, sweep/chirp, broadband noise and pulsed interference. Experimental results show that the deep learning models are effective in achieving high classification accuracy and offer reliable identification of jammers for real-time GNSS interference mitigation applications.