Deep Sequence-to-Sequence Models for GNSS Spoofing Detection

TL;DR

This paper introduces a deep learning framework using LSTM and Transformer models for real-time GNSS spoofing detection, achieving high accuracy with an error rate of 0.16% on a simulated dataset.

Contribution

It presents a novel data generation method for spoofing scenarios and applies advanced neural network architectures for online detection.

Findings

Deep learning models accurately detect spoofed signals.

Transformer-inspired models outperform other architectures.

Error rate of 0.16% achieved with the best model.

Abstract

We present a data generation framework designed to simulate spoofing attacks and randomly place attack scenarios worldwide. We apply deep neural network-based models for spoofing detection, utilizing Long Short-Term Memory networks and Transformer-inspired architectures. These models are specifically designed for online detection and are trained using the generated dataset. Our results demonstrate that deep learning models can accurately distinguish spoofed signals from genuine ones, achieving high detection performance. The best results are achieved by Transformer-inspired architectures with early fusion of the inputs resulting in an error rate of 0.16%.