journal article·2026-07-10Open access

Long-Term GNSS Satellite Clock Error Forecasting Using Inter-Satellite Comparison Data and RBF Neural Networks

Tieqiang Liu, Guocheng Wang, Li Liu, Yin Huang, Lintao Liu, Zhiwu Cai, Yu Xiao, Mingyuan Liu, Jianguo Wang

Applied Sciences· Volume 16· Issue 14

Abstract

Accurate long-term prediction of GNSS satellite clock errors is essential for autonomous navigation, real-time precise point positioning (PPP), and continuous positioning, navigation, and timing (PNT) services when real-time clock products are unavailable or delayed. However, conventional methods, such as quadratic polynomial fitting and Kalman filtering, have limited capability in modeling nonlinear and non-stationary clock behaviors over long prediction intervals, especially under abnormal onboard atomic clock conditions. To address this issue, an inter-satellite comparison data (ISCD)-based radial basis function neural network (RBFNN) model is proposed for long-term satellite clock error prediction. Through correlation analysis, reference satellite clocks closely related to the target satellite are selected, and both ISCD and satellite-ground comparison data are integrated to establish a nonlinear prediction model. Experiments using GPS and Galileo satellite clock datasets demonstrate that the proposed method significantly improves long-term prediction accuracy. For 180-day prediction, the proposed model reduces the RMSE by more than 60% for GPS satellites and approximately 99% for Galileo satellites with abnormal clock behavior compared with conventional methods. Rolling prediction experiments further verify the robustness and stability of the proposed model.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: GNSS positioning and interference · Advanced Frequency and Time Standards · Inertial Sensor and Navigation

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

State Key Laboratory of Geodesy and Earth’s Dynamics, Innovation Academy for Precision Measurement Science and Technology, Chinese Academy of Sciences, Wuhan 430071, China

University of Chinese Academy of Sciences, Beijing 100049, China

The 54th Research Institute of China Electronics Technology Group Corporation, Shijiazhuang 050081, China

Shenzhen Far-East Huaqiang Navigation and Position Co., Ltd., Shijiazhuang 050081, China

Beijing Satellite Navigation Center, Beijing 100094, China

Funding

National Natural Science Foundation of China · Major Scientific and Technological Mission Bureau of the Chinese Academy of Sciences · Foundation of State Key Laboratory of Precision Geodesy

References (34)

Show source-provided references
1. Clock bias prediction of navigation satellite based on BWO-CNN-BiGRU-attention model10.1007/s10291-024-01809-1 ↗
2. Highly accurate short-term prediction of BDS-3 satellite clock bias based on SSA-ANFIS integrated modeling10.1007/s10291-025-01864-2 ↗
3. Liu, J. (2008). GPS Satellite Navigation and Positioning Principle and Method, Science Press. [2nd ed.].
4. Long-term satellite clock bias prediction based on ARMA model
5. Xie, X. (2020). Precise Orbit and Clock Determination for BDS-3 Satellites Using Inter-Satellite Link Observations. [Ph.D. Thesis, Wuhan University].10.1007/s00190-020-01394-4 ↗
6. Inter-satellite clock offsets adjustment based on closed-loop residual detection of BDS inter-satellite link
7. Ai, Q., Maciuk, K., Lewinska, P., and Borowski, L. (2021). Characteristics of onefold clocks of GPS, Galileo, BeiDou and GLONASS systems. Sensors, 21.10.3390/s21072396 ↗
8. Orbit and clock analysis of BDS-3 satellites using inter-satellite link observations10.1007/s00190-020-01394-4 ↗
9. Sun, L., Huang, W., Gao, S., Li, W., Guo, X., and Yang, J. (2020). Joint timekeeping of navigation satellite constellation with inter-satellite links. Sensors, 20.10.3390/s20030670 ↗
10. Accuracy evaluation and improvement strategies of BeiDou broadcast clock error model
11. Huang, G., Cui, B., Zhang, Q., Fu, W., and Li, P. (2018). An improved predicted model for BDS ultra-rapid satellite clock offsets. Remote Sens., 10.10.3390/rs10010060 ↗
12. Long-term prediction of satellite clock offset based on artificial neural network10.12677/GST.2022.103013 ↗
13. A New Navigation Satellite Clock Bias Prediction Method Based on Modified Clock-bias Quadratic Polynomial Model
14. Application of grey model and autoregressive model combination in short-term clock prediction
15. Optimized Initial Condition of GM(1,1) Model and Its Application in GNSS Clock Offset Prediction
16. Davis, J., Bhattarai, S., and Ziebart, M. (2012, January 21–24). Development of a Kalman filter based GPS satellite clock time-offset prediction algorithm. Proceedings of the IEEE European Frequency and Time Forum, Geneva, Switzerland.10.1109/EFTF.2012.6502355 ↗
17. Huang, G. (2012). Research on Algorithm of Precise Clock Offset and Quality Evaluation of GNSS Satellite Clock. [Ph.D. Thesis, Chang’an University].
18. Influence of prediction model and modeling sequence length on short-term accuracy of clock bias
19. Research on wavelet neural network algorithm for satellite clock bias prediction
20. Research on Elman neural network algorithm to predict navigation satellite clock bias
21. Application of LSTM neural network in satellite clock prediction during satellite clock frequency deterioration
22. Satellite clock bias prediction based on Takagi-Sugeno fuzzy neural network
23. BP neural network for satellite clock bias prediction based on sparrow search algorithm
24. The application of radial basis function neural network in GPS satellite clock bias prediction
25. BDS-3 clock error prediction algorithm based on semi-parametric-SVM combined model
26. Clock bias prediction based on linear weighted combination Kalman filter
27. Determination of Covariance Matrix of Kalman Filter Used for Time Prediction of Atomic Clocks of Navigation Satellites
28. Accelerated gradient algorithm for RBF neural network10.1016/j.neucom.2021.02.009 ↗
29. Feng, Y. (2023). Research on BDS-3 Clock Difference Predictionby RBF Neural Network Based on Multipleoptimization Algorithms. [Ph.D. Thesis, Liaoning University of Engineering and Technology].
30. Satellite clock offset prediction using a combined model of GM and RBF neural network
31. Satellite Clock Bias Prediction Algorithm with Multi System Based on RBF Neural Network
32. Research of satellite clock error prediction based on RBF neural network and ARMA model
33. A new method for estimating the correlation of seismic waveforms based on the NTFT10.1093/gji/ggab047 ↗
34. Research of satellite clock error prediction based on neural networks