Enhancing Outer-Grid GNSS STEC Correction Accuracy Using Bias Removal and IPP Zenith Angle Regression

Young-Woo Jeon*, Kwan-Dong Park

Ionospheric delay is a dominant error source in precise positioning using Global Navigation Satellite System (GNSS). Recently, grid-based Slant Total Electron Content (STEC) corrections without the conventional single-layer assumption have been investigated. However, degradation in the ionosphere’s physical consistency during Differential Code Bias (DCB) correction and reduced interpolation stability near network boundaries remain major limitations. This study generates bias-corrected STEC preserving ionospheric physical characteristics and improves outer-grid correction accuracy using a K-nearest neighbor (K-NN)-based interpolation model incorporating the zenith angle of the Ionospheric Pierce Point (IPP) and adaptive use of additional neighboring observations. Satellite-specific STEC corrections were derived using CODE’s Observable-Specific Bias (OSB) products within an Extended Kalman Filter (EKF) framework. Cross-validation results indicate accuracy comparable to conventional linear interpolation in inner-grid regions, whereas in outer-grid regions, the proposed method achieved an RMSE of 0.35 m, corresponding to a 34% improvement over conventional linear interpolation.
Keywords: GNSS, DCB, ionospheric corrections, K-NN regression

Speaker 전영우* 인하대학교