Self-Supervised Federated GNSS Spoofing Detection with Opportunistic Data

TL;DR

This work tackles GNSS spoofing detection by combining self-supervised learning with federated learning to avoid labeling cost and protect location privacy. Edge devices train LSTM-based anomaly detectors on locally generated spoofing-deviation labels, while a cloud server aggregates updates via FedAvg in an iterative cycle. The approach outperforms a position-based baseline and approaches centralized-training performance, while enabling privacy-preserving collaboration across devices and traces. Experiments on a real-world Jammertest 2024 dataset demonstrate robust generalization across devices and traces and reveal practical considerations for deployment. The method offers a scalable, privacy-conscious pathway for real-time GNSS security in distributed mobile ecosystems.

Abstract

Global navigation satellite systems (GNSS) are vulnerable to spoofing attacks, with adversarial signals manipulating the location or time information of receivers, potentially causing severe disruptions. The task of discerning the spoofing signals from benign ones is naturally relevant for machine learning, thus recent interest in applying it for detection. While deep learning-based methods are promising, they require extensive labeled datasets, consume significant computational resources, and raise privacy concerns due to the sensitive nature of position data. This is why this paper proposes a self-supervised federated learning framework for GNSS spoofing detection. It consists of a cloud server and local mobile platforms. Each mobile platform employs a self-supervised anomaly detector using long short-term memory (LSTM) networks. Labels for training are generated locally through a spoofing-deviation prediction algorithm, ensuring privacy. Local models are trained independently, and only their parameters are uploaded to the cloud server, which aggregates them into a global model using FedAvg. The updated global model is then distributed back to the mobile platforms and trained iteratively. The evaluation shows that our self-supervised federated learning framework outperforms position-based and deep learning-based methods in detecting spoofing attacks while preserving data privacy.

Figures (11)

• Figure 1: System and adversary model illustration.
• Figure 2: Overview of self-supervised federated spoofing detection.
• Figure 3: Jammertest main test area (right) and mounted smartphones in a vehicle (left).
• Figure 4: curves of the centralized and federated self-supervised detection, and the position-based detection.
• Figure 5: curves of the proposed self-supervised detection for each smartphone based on their local data.