Improving prediction performance of GPS satellite clock bias based on wavelet neural network
- 528 Downloads
As one of the IGS ultra-rapid predicted (IGU-P) products, the orbit precision has been remarkably improved since late 2007. However, because satellite atomic clocks in space show complicated time–frequency characteristics and are easily influenced by many external factors such as temperature and environment, the IGU-P clock products have not shown sufficient high-quality prediction performance. An improved prediction model is proposed in order to enhance the prediction performance of satellite clock bias (SCB) by employing a wavelet neural network (WNN) model based on the data characteristic of SCB. Specifically, two SCB values of adjacent epoch subtract each other to get the corresponding single difference sequence of SCB, and then, the sequence is preprocessed through using the preprocessing method designed for the single difference sequence. The subsequent step is to model the WNN based on the preprocessed sequence. After the WNN model is determined, the next single difference values at the back of the modeling sequence are predicted. Lastly, the predicted single difference values are restored to the corresponding predicted SCB values. The simulation results have shown that the proposed prediction principle based on the single difference sequence of SCB can make the WNN model simple in architecture and the predicting precision higher than that of the general SCB prediction modeling. The designed preprocessing method specific to the single difference of SCB is able to further improve the prediction performance of the WNN model by reducing the effect from outliers. The proposed SCB prediction model outperforms the IGU-P solutions at least on a daily basis. Specifically, the average prediction precisions for 6, 12 and 24 h based on the proposed model have improved by about 13.53, 31.56 and 49.46 % compared with the IGU-P clock products, and the corresponding average prediction stabilities for 12 and 24 h have increased by about 13.89 and 27.22 %, while the average prediction stability of 6 h is nearly equal.
KeywordsSatellite clock bias Prediction model Data preprocessing Wavelet neural network IGU
The authors thank the reviewers for their beneficial comments and suggestions. This study has been supported by National Natural Science Foundation of China (Grant Nos. 41274015 and U1431115), Chinese High-tech R&D (863) Program (Grant No. 2013AA122501), Open Research Foundation of State Key Laboratory of Geo-information Engineering (Grant No. SKLGIE2015-M-1-6).
- Davis J, Bhattarai S, Ziebart M (2012) Development of a Kalman filter based GPS satellite clock time-offset prediction algorithm. European Frequency and Time Forum (EFTF), pp 152–156Google Scholar
- Ray J, Senior K (2003) IGS/BIPM Pilot Project: GPS carrier phase for time/frequency transfer and time scale formation. 4th international time scale algorithms symposium at BIPM, Sevres, France, 18–19 March 2002. Metrologia 40(3):270–288; Erratum 40(4): 205Google Scholar
- Riley W (2002) The calculation of time domain frequency stability. Hamilton Technical Services, BeaufortGoogle Scholar
- Riley W (2003) Techniques for frequency stability analysis. In: IEEE international frequency control symposium, Tampa, FL (vol 4)Google Scholar
- Riley W (2007) Handbook of frequency stability analysis. Hamilton Technical Services, BeaufortGoogle Scholar
- Senior K, Ray J (2001) Accuracy and precision of carrier phase clock estimates. In: Proceedings of 33rd precise time and time interval applications and planning meeting, pp 199–217Google Scholar
- Senior K, Koppang P, Matsakis D, Ray J (2001) Developing an IGS time scale. In: Proceedings of 2001 IEEE international frequency control symposium, and PDA exhibition, pp 211–218Google Scholar
- Wang JG (2010) Research on Time Comparison based on GPS precise point positioning and atomic clock prediction. Dissertation, Graduate School of CAS, BeijingGoogle Scholar
- Wang YP, Lu ZP, Chen ZS, Cui Y (2013) Research the algorithm of wavelet neural network to predict satellite clock bias. Acta Geodaetica Cartogr Sin 42(3):323–330Google Scholar
- Xu G (2007) GPS: theory, algorithms and applications. Springer Science & Business Media, BerlinGoogle Scholar
- Zheng ZY, Lu XS, Chen YQ (2008) Improved grey model and application in real-time GPS satellite clock bias prediction. In: Proceedings of 4th international conference on natural computation, pp 419–423Google Scholar