Electrical and Computer Engineering ETDs

Publication Date

Spring 5-16-2026

Abstract

Global Navigation Satellite Systems (GNSS) are vulnerable to spoofing attacks that can mislead receivers with counterfeit signals. Traditional detection techniques, such as antenna-based, encryption based, and signal processing approaches, often face limitations in adaptability, computational cost, or reliance on predefined thresholds. Supervised machine learning models, while powerful, require large labeled datasets and struggle to generalize to unseen spoofing scenarios. In this work, we propose a self-supervised spoofing detection framework based on Adaptive Sparse Gaussian Processes (ASGP). The method predicts incoming GNSS features using past observations and identifies spoofing as anomalous deviations in the prediction residuals. Unlike supervised approaches, ASGP adapts online to dynamic, non-stationary environments while maintaining configurable computational complexity. We evaluate the approach using the OAKBAT dataset under various time push and position-push attack scenarios. Results demonstrate that ASGP achieves superior detection performance compared to Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN), and Support Vector Machine (SVM) models when applied in a self-supervised setting. The proposed method provides an efficient, flexible, and label-free alternative for robust spoofing detection in practical GNSS applications.

Keywords

Adaptive Sparse Gaussian Processes, GPS, Self-Supervised Learning, Spoofing Detection.

Document Type

Dissertation

Language

English

Degree Name

Computer Engineering

Level of Degree

Doctoral

Department Name

Electrical and Computer Engineering

First Committee Member (Chair)

Dr. Manel Martinez-Ramon

Second Committee Member

Dr. Ramiro Jordan

Third Committee Member

Dr. Christos Christodoulou

Fourth Committee Member

Dr. Trilce Estrada

Fifth Committee Member

Mr. Brian Zufelt

Download

Share

COinS