LEO satellite clock analysis and prediction for positioning applications

The positioning service aided by low Earth orbit (LEO) mega-constellations has become a hot topic in recent years. To achieve precise positioning, accuracy of the LEO clocks is important for single-receiver users. bridge gap between applicable time clock products and need to be predicted over certain period depending on sampling interval products. This study discusses prediction errors periods from 10 s 1 h two typical types, i.e. ultra-stable oscillator (USO) oven-controlled crystal (OCXO). based GNSS-determined estimates, where stability related GNSS estimation errors, behaviors oscillators themselves, systematic effects environment relativistic effects, reference. Based real data analysis, different types are simulated under conditions, model considering proposed. Compared simple polynomial fitting usually applied, proposed can significantly reduce about 40%-70% simulations 5%-30% containing miss-modeled effects. For both short-term min would result root mean square error (RMSE) few centimeters when using very stable RMSE amounts 0.1 m, real-time reference national center space studies (CNES) was used. long-term h, could range below m meters USOs, complexity OCXOs, lead larger with an m.