Enhanced NRTK positioning based on LEO satellite undifferenced observations at the user side

Abstract The high-speed movement of LEO (low Earth orbit) satellites helps improve the GNSS (Global navigation satellite system) satellite geometry and reduce the correlation between observation equations across epochs, thereby shortening the initialization time and enabling rapid positioning. An network real-time kinematic enhancement algorithm based on undifferenced observations of LEO satellites is proposed in this paper. The approach integrates GNSS virtual observations generated from a medium-to-long-range reference station network with the LEO satellite signals received by the user station. Experimental data show that after adding LEO satellites, over half of the positioning results converge within 10 min. The average time to first fix is reduced to 5.6 min, and the average success rate of fixing increases from 82% to 96%, with a relative improvement of 18%. In terms of positioning accuracy, the root mean square values of the positioning errors in the East, North, and Up directions are reduced to 1.5, 1.9, and 3.8 cm, respectively, representing improvements of 58%, 50%, and 39% compared to the case without LEO satellites.