Exploiting Signals of Opportunity for High-Precision PNT: Optimized Bayesian Estimation Across 5G and LEO Networks

The rapid advancement of autonomous technologies and highly dynamic platforms demands continuous, high-precision positioning, navigation, and timing (PNT) systems. While traditional global navigation satellite systems (GNSS) degrade severely in deep urban canyons, emerging terrestrial 5G millimeter-wave and non-terrestrial low Earth orbit (LEO) satellite signals of opportunity offer highly promising alternatives. However, the practical realization of a seamless, hybrid navigation architecture is currently hindered by profound estimation and geometric bottlenecks. In the terrestrial domain, the measurement quality of 5G signals fluctuates drastically due to rapidly changing spatial geometry, varying signal-to-noise ratios (SNR), and kinematic Doppler shifts, frequently leading to dangerous filter overconfidence in standard Bayesian estimators. Conversely, in the non-terrestrial domain, standalone LEO Doppler positioning suffers from fundamentally weak vertical observability, while multi-receiver differential architectures are severely corrupted by the unmodeled, asynchronous clock drifts inherent in LEO communication constellations. This thesis makes four main contributions. First, a data-driven covariance adaptation framework is developed for terrestrial 5G positioning, enabling Bayesian navigation filters to dynamically adjust measurement trust based on theoretically predicted performance bounds and preventing filter overconfidence. Second, a digital terrain model (DTM)-aided LEO Doppler positioning architecture is proposed to resolve the inherent altitude ambiguity of single-receiver systems and enable accurate three-dimensional positioning. Third, the Cramér–Rao lower bound (CRLB) for multi-receiver LEO Doppler positioning is derived under spatially correlated atmospheric conditions, providing a theoretical benchmark for evaluating differential architectures. Fourth, a base-station-aided differential LEO framework is developed to explicitly estimate and mitigate asynchronous satellite clock drifts. To validate the theoretical and architectural contributions of this thesis, the proposed methodologies were evaluated using a combination of real-world vehicular data and high-fidelity simulations. Specifically, empirical driving trajectories collected in downtown Toronto and Kingston, Ontario, were integrated with advanced 5G and LEO propagation models to rigorously benchmark the developed algorithms.

Department dc:contributor.department

Electrical and Computer Engineering

Year dc:date.issued

2026

Author dc:creator

• Bader, Qamar Muneer M A

Advisor dc:contributor.supervisor

• Noureldin, Aboelmagd

• LEO
• PNT
• 5G
• Sensor Fusion
• Kalman Filtering

Statement dc:rights

• Attribution-NonCommercial-NoDerivatives 4.0 International

Licence dc:rights.uri

• http://creativecommons.org/licenses/by-nc-nd/4.0/

Language dc:language.iso

eng

Handle dc:identifier.uri

https://hdl.handle.net/1974/36356

OAI identifier oai:identifier

oai:queensu.scholaris.ca:1974/36356