Satellite-based navigation systems underpin nearly every modern infrastructure, and thus their limitations are well known. Standard global navigation satellite systems (GNSS), such as Galileo and GPS, struggle with signal loss in obstructed environments such as indoors, dense urban canyons, or remote areas. M.Sc. Kaan Çelikbilek’s dissertation has explored how emerging low-earth orbit (LEO) satellite constellations could overcome these weaknesses and even outperform existing systems.
The dissertation research introduces a set of novel constellation designs tailored specifically for positioning, navigation, and timing (PNT) services. These designs were developed using advanced multi-objective optimization methods, balancing performance, cost-efficiency, and orbital stability. Compared to traditional GNSS architectures, these LEO-PNT constellations can deliver faster signals with greater strength and improved geometric coverage, particularly valuable in environments where the signal degradation has traditionally been a barrier towards achieving satisfactory performance.
LEO-PNT systems represent a viable path toward improved standalone navigation capabilities
Performance evaluations across a variety of scenarios, including challenging indoor conditions, indicate that optimized LEO-PNT constellations not only match, but surpass current GNSS in accuracy and reliability. One standout result demonstrates that integrating two LEO constellations (or one LEO with an existing GNSS) can boost the overall signal performance by more than 100%, depending on the configuration.
The findings also challenge the assumption that large-scale, single-shell satellite networks are optimal. Instead, multi-shell LEO architecture have emerged as more efficient, with three-shell configurations offering a near-ideal trade-off between complexity and performance. These optimized setups exhibited strong potential to maintain high-quality navigation services in both urban and rural environments.
Most importantly, the study suggests that LEO-PNT systems are not just enhancements to existing infrastructure, but they represent a viable path toward standalone navigation capabilities, laying the groundwork for future technologies such as autonomous transport, emergency services in hard-to-reach zones, and next-generation geolocation services.
Beyond current applications, the research outlines clear paths for future development. These include hybrid positioning techniques combining Doppler and code-based methods, security improvements against jamming and spoofing, and the integration of machine learning for dynamic signal optimization in real time. The study strongly implies that the case for transitioning to LEO-driven navigation is no longer speculative, it’s increasingly practical, and soon, perhaps, necessary.
Public defence on Friday 23 May
The doctoral dissertation of M. Sc. (Eng.) Kaan Çelikbilek in the field of electrical engineering titled Simulation, Optimization, and Comparisons Among LEO-PNT Constellations will be publicly examined at the Faculty of Information Technology and Communication Sciences at Tampere University on Friday, 23 May 2025, at 12.00, at the Tietotalo building of Hervanta campus, in the TB109 auditorium (address: Korkeakoulunkatu 1, 33720 Tampere).
The Opponent will be Assistant Professor Rafael Berkvens from the Faculty of Applied Engineering at the University of Antwerp. The Custos will be Professor Elena Simona Lohan from the Faculty of Information Technology and Communication Sciences of Tampere University.