Skip to content
Conference

Cross-Domain PRN-Level GNSS Spoofing Detection Using Few-Shot Reptile Meta-Learning

Oct 2026 · International Conference on Localization and Global Navigation Satellite System · pp. 1-7 · 0 citations · 31 references

Abstract

Positioning, navigation, and timing (PNT) solutions based on Global Navigation Satellite Systems (GNSS) are crucial for many applications nowadays, from transport and energy networks to mobile communications and finance. However, spoofing attacks, where counterfeit signals mimic authentic ones, are a critical threat to satellite-based PNT. Numerous methods have recently been proposed in the literature to detect spoofing, showing promising results, but their robustness under cross-scenario and cross-dataset conditions remains insufficiently investigated. The machine learning (ML) and deep learning models used in prior studies typically require a large amount of training data, which increases the computational cost. This paper implements a Reptile-based meta-learning framework for GNSS spoofing detection using tracking features and few-shot adaptation to enhance generalization under cross-scenario and cross-dataset conditions. The methodology relies on publicly available datasets including the Texas Spoofing Test Battery (TEXBAT), the Oak Ridge Spoofing and Interference Test Battery (OAKBAT) and the Finnish Geospatial Research Institute spoofing dataset (FGI-SpoofRepo). Experimental results confirm that the implemented Reptile-based framework provides a more accurate and computationally efficient solution than classical ML methods and some of the meta-learning models.

View source

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