Skip to content
Conference Open access

GNSS Jamming and Spoofing Detection Using NMEA Data

Oct 2026 · International Conference on Localization and Global Navigation Satellite System · pp. 1-7 · 0 citations · 22 references

Abstract

Global Navigation Satellite Systems (GNSS) are widely used for Positioning, Navigation, and Timing (PNT) in many critical applications. However, these systems are vulnerable to intentional interference, including jamming and spoofing attacks, which degrade performance and compromise the reliability of location-based services. Detecting such disruptions remains challenging, due to the limited availability of realistic and publicly accessible measurement data. The Jammertest campaign is one of the few legal sources of real GNSS interference data, including controlled jamming and spoofing scenarios. Despite multiple annual campaigns, access remains limited, restricting research on robust detection methods. In this paper, we investigate Machine Learning (ML) and Deep Learning (DL) approaches for classifying GNSS signal conditions into normal, jammed, and spoofed states using real measurements from Jammertest 2025. We compare tree-based models (Random Forest, XGBoost, LightGBM) and sequence models (LSTM, BiLSTM, GRU) under a cross-testing strategy evaluating generalization to unseen interference. All models achieve near-perfect in-distribution accuracy, while on cross-test, sequence models generalize better, with BiLSTM reaching 0.93 accuracy, 0.91 macro F1, and 0.89 macro recall.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.