Optimizing for Real-Time Performance through Determining the Optimal Robust Filter Configuration for Multi-GNSS Smartphone RTK

Yuting Gao, Shengqi Wang, Yang Jiang, Yang Gao

Peer Reviewed

Abstract:	In recent years, the rapid evolution of multi-constellation Global Navigation Satellite Systems (GNSS) and the fast integration of multi-frequency, multi-system GNSS receivers into consumer smartphones have collectively driven the development of high-precision positioning capabilities into the everyday lives of the public. However, real-time kinematic (RTK) positioning on smartphones remains hindered by persistent limitations in solution stability and reliability. These limitations stem primarily from severe multipath and non-line-of-sight (NLOS) signal distortions—made worse by the not ideal radiation characteristics and placement constraints of embedded antennas—as well as inherent computational and power constraints of mobile devices. To overcome these challenges, this work introduces a lightweight, real-time RTK positioning framework designed specifically for unmodified commercial smartphones, achieving consistent meter-level accuracy under dynamic rural conditions. We further conduct a systematic evaluation of two robust estimation strategies: GNSS Epoch Difference–Solution Separation (GED-SS) and Single Point Positioning–Solution Separation (SPP-SS). Experiments conducted across diverse rural canyon scenarios confirm that both methods preserve low-latency, computationally efficient operation while delivering significant improvements in positioning continuity and resilience to signal degradation compared with conventional approaches. Moreover, we establish a standardized, open experimental framework enabling fair, reproducible benchmarking of three robust filtering methodologies—including the baseline Robust Kalman Filter. Through controlled parameter sweeps and performance profiling, this study quantifies the sensitivity of positioning accuracy and availability to key robustness tuning parameters, thereby offering empirically grounded design principles and theoretical insights for developing resilient GNSS positioning algorithms in GNSS-challenged rural environments.
Published in:	Proceedings of the ION 2026 Pacific PNT Meeting April 13 - 16, 2026 Hilton Waikiki Beach Honolulu, Hawaii
Pages:	140 - 155
Cite this article:	Gao, Yuting, Wang, Shengqi, Jiang, Yang, Gao, Yang, "Optimizing for Real-Time Performance through Determining the Optimal Robust Filter Configuration for Multi-GNSS Smartphone RTK," Proceedings of the ION 2026 Pacific PNT Meeting, Honolulu, Hawaii, April 2026, pp. 140-155. https://doi.org/10.33012/2026.20592
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In