Integrity and attack-resilience of GPS-based positioning and timing: a Bayesian and measurement fusion approach

Robust Position, Velocity, and Timing (PVT) are essential for the safe operations of critical infrastructure sectors, such as transportation systems and power grids. Different transportation systems, both human-operated and autonomous vehicles, navigate using accurate position and velocity information. On the other hand, precise timing is crucial for various economic activities worldwide, such as banking, stock markets, and the power grid. GPS serves as a backbone for many state-of-the-art applications related to these crucial infrastructures. GPS provides sub-microsecond accurate timing and meter level of accurate positioning. It has global coverage and is free for all users. The GPS positioning and timing service has some limitations. The positioning accuracy degrades in urban environments due to tall structures that block and reflect satellite signals. Degraded positioning is not safe for the operation of autonomously driving vehicles. Furthermore, GPS signals are susceptible to external attacks due to their low signal power and unencrypted signal structures. Researchers have shown that GPS Spoofing Attacks (GSAs) are feasible, and GSA for timing is able to alter timing without modifying the positioning solution. Such attacks create unsafe operating conditions for the modern power grid, which will use GPS timing for monitoring the wide-area network. The contribution of this work is to develop algorithms to mitigate the above limitations. We develop Bayesian algorithms that utilize multiple sensors and receivers. For improving positioning, first, we design an adaptive filter based on Bayesian algorithms to augment GPS with the additional vision sensor. Second, we develop an integrity monitoring algorithm for Direct Positioning (DP), which is an advanced GPS receiver architecture that directly works on the position domain and is robust to signal blockage and multipath effects. To monitor integrity, we estimate vertical protection levels using a Bayesian approach. We further generate GPS datasets simulating open, semi-urban, and urban environments for validating DP with multiple receivers. For mitigating GSAs for timing, we design static and dynamic state estimators for the power grid. The static state estimator utilizes measurement residuals to correct power grid states. In the dynamic state estimator, we fuse GPS and power grid measurements to provide resiliency against GSAs. We create a virtual power grid testbed and generate datasets for a power grid network under different GSAs. These are the first datasets that contain both power grid and GPS measurements under GSAs, and we make them openly available. Our estimators are validated on various power grid networks and on the generated datasets.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Aerospace Engineering

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2021

Author dc:creator

• Chauhan, Shubhendra Vikram Singh

Contributors dc:contributor

• Gao, Grace
• Bretl, Timothy
• Tran, Huy
• Sauer, Peter

• Adaptive Filter
• Image Localization
• Global Positioning System (GPS)
• Protection Levels
• Direct Positioning
• Bayesian approach
• State Estimation
• Phasor Measurement Unit (PMU)
• Measurement Fusion
• GPS Spoofing Attacks
• GPS-PMU Integrated Datasets

Statement dc:rights

• Copyright 2020 Shubhendra Chauhan

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/109364

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/109364