Improving multipath extraction in PPP-RTK for high-precision dynamic deformation monitoring
Year: 2025
Type: article
Abstract: Multipath error is one of the most critical error sources in GNSS precise-positioning applications. Extracting accurate observational residuals is a fundamental and challenging task for constructing multipath corrections in a Precision Point Positioning Real-Time Kinematics (PPP-RTK) model. We proposed a novel multipath extraction method in UnDifferenced and UnCombined (UDUC) PPP-RTK framework. Instead of extracting multipath directly from UD residuals, our method first extracts inter-satellite Single-Difference (SD) residuals and then converts them to UD residuals. Additionally, we replace the fixed average position parameters used in traditional multipath modeling with trends that reflect the true deformation at the station, accommodating both stable and dynamic motion scenarios. Experimental results in real landslide scenarios demonstrate that the proposed method reduces the impact of receiver clock offset and other satellite ambiguity parameters on residual extraction and is suitable for both stable and dynamic motion conditions. Results from two reference networks with average spacings of 105 km and 214 km demonstrate that the proposed method achieved positioning accuracies of 1.1 cm, 1.2 cm, 3.2 cm in the east, north, up directions, and 1.0 cm, 1.3 cm, 3.3 cm, respectively. Compared to the traditional multipath extraction method that uses averaged position parameters, these results represent improvements of 15%, 14%, 20%, and 37%, 28%, 23%, respectively.
Source: Geo-spatial Information Science
Authors: Xinrui Li, Bao Shu, Xuanyu Qu, Qin Zhang, Yunqing Tian, Guanwen Huang, Li Wang
Institutions: Chang'an University, Hong Kong Polytechnic University
Language: English
Subfield: Aerospace Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $1,993
Award: G1915253765
OpenAlex ID: W4413351729
