Ionospheric nonlinear interpolation model for Mid- and Low-latitude network RTK during solar maxima

Abstract

As the Sun approaches the peak of its 25th activity cycle, intensified ionospheric spatial gradients near the equatorial ionization anomaly (EIA) crest pose critical challenges to GNSS positioning accuracy, particularly in low-latitude regions. Traditional network RTK systems, which rely on linear interpolation models (LIMs) to approximate ionospheric correlations between reference stations, inadequately resolve nonlinear spatial gradients along ionospheric pierce point (IPP) trajectories, a key source of residual errors in double-differenced ionospheric (DDI) delays. To address this limitation, we introduce a Nonlinear Interpolation Model (NIM) that explicitly incorporates satellite-specific gradients along IPP trajectories. By dynamically detrending spatially nonlinear ionospheric terms, NIM improves ionospheric delay interpolation accuracy. Evaluations across mid-latitude and low-latitude networks show NIM reduces DDI interpolation errors by 30–40% compared to LIMs. Statistical analyses under diverse ionospheric conditions highlight NIM’s enhanced error distribution characteristics, particularly during sunset and post-sunset transitions when gradients peak. Notably, during the extreme May 2024 geomagnetic storm, NIM achieved below 2 cm RMS positioning accuracy in Hong Kong, a region historically prone to large ionospheric gradients. These improvements translate to measurable gains: a 10% higher ambiguity resolution success rate, 30% faster convergence times, and horizontal/vertical positioning precision of 1.1/3.8 cm. By integrating IPP trajectory gradients into spatial modeling, NIM provides a scalable framework for robust RTK operations in gradient-prone regions. This advancement supports reliable centimeter-level positioning during solar maxima.