Robust Positioning Algorithm for Challenging Environments by Hybridizing GNSS and 5G TN/NTN

Enrique Domínguez-Tijero

Abstract:	Reliable positioning in dense urban environments remains challenging for Global Navigation Satellite Systems (GNSS) due to frequent signal blockage, severe multipath, and non-Gaussian, time-correlated errors. This paper proposes a robust hybrid navigation algorithm that fuses GNSS with 5G terrestrial networks (TN) and low-Earth-orbit (LEO) non-terrestrial networks (NTN) within a fixed-lag Factor Graph Optimization (FGO) framework. The method jointly estimates user states—position, velocity, clock bias and drift—and models non-compensated 5G TN/NTN synchronization offsets. We keep a sliding window of recent states and marginalize old ones to keep computation bounded. At each step we apply ?² gating and a Huber loss to individual measurements, and we add window-level tests (grouped ?², trend and change-point detector tests) to catch NLoS errors. The factor graph models pseudorange and Doppler and, when dual-frequency is available, can also estimate satellite and ionospheric biases. This helps handle correlated and heavy-tailed errors in a consistent way. We validate the approach with real GNSS data collected in deep-urban conditions at 1 Hz, augmented by realistic 5G TN and 5G NTN LEO ranging/Doppler scenarios. Across these trials, FGO outperforms classical EKF baselines, and the inclusion of 5G TN/NTN factors further improves availability and reduces horizontal error tails during GNSS masking and multipath. The results demonstrate that leveraging the complementary geometry and failure modes of GNSS and 5G—together with estimationlevel robustness—delivers a practical accuracy and resilience uplift without adding new user hardware. FGO commonly leverages inertial propagation, and the proposed framework can be extended to additional sensing (e.g., IMU), offering a scalable pathway to robust urban navigation for those applications that have them available. However, in this study we evaluate a radio-only FGO (no INS) to assess the benefits of smoothing and robust estimation when fusing GNSS with 5G TN/NTN. This choice broadens applicability to platforms that lack high-quality IMUs or where power, cost, or integration constraints preclude inertial usage. Although this study focuses on a radio-only FGO (no INS), the proposed FGO architecture can be easily extended to employ IMU measurements by introducing IMU preintegration factors into the factor-graph pipelines.
Published in:	Proceedings of the 38th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2025) September 8 - 12, 2025 Hilton Baltimore Inner Harbor Baltimore, Maryland
Pages:	2739 - 2753
Cite this article:	Domínguez-Tijero, Enrique, "Robust Positioning Algorithm for Challenging Environments by Hybridizing GNSS and 5G TN/NTN," Proceedings of the 38th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2025), Baltimore, Maryland, September 2025, pp. 2739-2753. https://doi.org/10.33012/2025.20455
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In