Developing an Innovative High-precision Approach to Predict Medium-term and Long-term Satellite Clock Bias

doi:10.11947/j.JGGS.2023.0104

Abstract

Abstract:

A new prediction method based on the nonlinear autoregressive model is proposed to improve the accuracy of medium-term and long-term predictions of Satellite Clock Bias (SCB). Forecast experiments for three time periods were implemented based on the precision SCB published on the International GNSS Server (IGS) server. The results show that the medium-term and long-term prediction accuracy of the proposed approach is significantly better compared to other traditional models, with the training time being much shorter than the wavelet neural network model.

Key words: Satellite Clock Bias (SCB); Median Absolute Deviation (MAD); wavelet threshold; nonlinear autoregressive model

Xu WANG,Hongzhou CHAI. Developing an Innovative High-precision Approach to Predict Medium-term and Long-term Satellite Clock Bias[J]. Journal of Geodesy and Geoinformation Science, 2023, 6(1): 47-58.

Figures/Tables 11

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Tab.1

Fig.7

Tab.2

Tab.3

Tab.4

References 30

[1]	JIANG Chen, ZHANG Shubi, CAO Yizhi, et al. A robust fault detection algorithm for the GNSS/INS integrated navigation systems[J]. Journal of Geodesy and Geoinformation Science, 2020, 3(1): 12-24. DOI: 10.11947/j.JGGS.2020.0102. doi: 10.11947/j.JGGS.2020.0102
[2]	WANG Yupu, LYU Zhiping, CHEN Zhengsheng, et al. Research on the algorithm of wavelet neural network to predict satellite clock bias[J]. Acta Geodaetica et Cartographica Sinica, 2013, 42(3): 323-330.
[3]	WANG Yupu, LU Zhiping, QU Yunying, et al. Improving prediction performance of GPS satellite clock bias based on wavelet neural network[J]. GPS Solutions, 2017, 21(2): 523-534. doi: 10.1007/s10291-016-0543-z
[4]	WANG Yupu. Research on modeling and prediction of the satellite clock bias and performance evaluation of GNSS satellite clocks[J]. Acta Geodaetica et Cartographica Sinica, 2018, 47(7): 1026. DOI: 10.11947/j.AGCS.2018.20170467. doi: 10.11947/j.AGCS.2018.20170467
[5]	ZHOU Peiyuan, DU Lan, LU Yu, et al. Periodic variations of BeiDou satellite clock offsets derived from multi-satellite orbit determination[J]. Acta Geodaetica et Cartographica Sinica, 2015, 44(12): 1299-1306. DOI: 10.11947/j.AGCS.2015.20150183. doi: 10.11947/j.AGCS.2015.20150183
[6]	ZHANG Qianqian, HAN Songhui, DU Lan, et al. Bayesian methods for outliers detection and estimation in clock offset measurements of satellite-ground time transfer[J]. Geomatics and Information Science of Wuhan University, 2016, 41(6): 772-777.
[7]	CASEIRO L M A, MENDES A M S. Real-time IGBT open-circuit fault diagnosis in three-level neutral-point-clamped voltage-source rectifiers based on instant voltage error[J]. IEEE Transactions on Industrial Electronics, 2015, 62(3): 1669-1678. doi: 10.1109/TIE.2014.2341558
[8]	WU Jing. Study on detection of GPS clock jump using median absolute deviation[J]. Science of Surveying and Mapping, 2015, 40(6): 36-41.
[9]	YUAN Kaiming, SHU Naiqiu, SUN Yunlian, et al. Wavelet denoising based on threshold optimization method[J]. Engineering Journal of Wuhan University, 2015, 48(1): 74-80.
[10]	LI Kejun, GAO Jinyao, SONG Jianhua, et al. Forecast of short-term metro passenger flow based on NARNN model[J]. Journal of Wuhan University of Technology (Transportation Science & Engineering), 2020, 44(1): 103-107.
[11]	WANG Xu, CHAI Hongzhou, WANG Chang. T-S fuzzy neural network to predict satellite clock bias[J]. Acta Geodaetica et Cartographica Sinica, 2020, 49(5): 580-588. DOI: 10.11947/j.AGCS.2020.20190156. doi: 10.11947/j.AGCS.2020.20190156
[12]	LIANG Yueji, REN Chao, YANG Xiufa, et al. Grey model based on first difference in the application of the satellite clock bias prediction[J]. Acta Astronomica Sinica, 2015, 56(3): 264-277.
[13]	JONSSON P, EKLUNDH L. Seasonality extraction by function fitting to time-series of satellite sensor data[J]. IEEE Transactions on Geoscience and Remote Sensing, 2002, 40(8): 1824-1832. doi: 10.1109/TGRS.2002.802519
[14]	WANG Jigang. Research on time comparison based on GPS precise point positioning and atomic clock prediction[D]. Xi’an: National Time Service Center, Chinese Academy of Sciences, 2010.
[15]	ZHU Xiangwei, XIAO Hua, YONG Shaowei, et al. The Kalman algorithm used for satellite clock offset prediction and its performance analysis[J]. Journal of Astronautics, 2008, 29(3):966-970, 1052.
[16]	ZHAO Liang, LAN Xiaoqi, SHENG Jianyue. Application of ARIMA model in satellite clock error forecasting[J]. Journal of Water Resources and Architectural Engineering, 2012, 10(1): 135-137.
[17]	WANG Yupu, LV Zhiping, CHEN Zhengsheng, et al. A new data preprocessing method for satellite clock bias and its application in WNN to predict medium-term and long-term clock bias[J]. Geomatics and Information Science of Wuhan University, 2016, 41(3): 373-379.
[18]	WANG Xu, CHAI Hongzhou, WANG Chang, et al. Improved wavelet neural network based on change rate to predict satellite clock bias[J]. Survey Review, 2021, 53(379): 325-334.DOI: 10.1080/00396265.2020.1758999 doi: 10.1080/00396265.2020.1758999
[19]	SHAN R, SHI S, LIU W, et al. Thesis on combination stock forecasting model with wavelet denoising and optimized arima method[J]. ICIC Express Letters, 2014, 8(8): 2315-2320.
[20]	XIAO Shungen, ZHANG Zexiong, SONG Mengmeng. Fault diagnosis method of gear based on lifting wavelet packet and combined optimization BP neural network[J]. Vibroengineering PROCEDIA, 2019, 29: 18-23. doi: 10.21595/vp
[21]	ZHANG Zhetao, ZHU Jianjun, KUANG Cuilin, et al. Multi-threshold wavelet packet de-noising method and its application in deformation analysis[J]. Acta Geodaetica et Cartographica Sinica, 2014, 43(1): 13-20. DOI: 10.13485/j.cnki.11-2089.2014.0003. doi: 10.13485/j.cnki.11-2089.2014.0003
[22]	SESIA I, CANTONI E, CERNIGLIARO A, et al. An efficient and configurable preprocessing algorithm to improve stability analysis[J]. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, 2016, 63(4):575-581. doi: 10.1109/TUFFC.2015.2496280
[23]	GUO Chengjun, TENG Yunlong. Performance analysis of satellite clock bias based on wavelet analysis and neural network[J]. Acta Astronomica Sinica, 2010, 51(4): 395-403.
[24]	WANG Xu, LIU Wensheng, WANG Chang. Processing of monitoring data of building deformation based on wavelet threshold denoising[J]. Engineering of Surveying and Mapping, 2011, 20(1): 44-46.
[25]	TAO Ke, ZHU Jianjun. A hybrid indicator for determining the best decomposition scale of wavelet denoising[J]. Acta Geodaetica et Cartographica Sinica, 2012, 41(5): 749-755.
[26]	YUE Xinzheng, LI Leimin, SUN Fei. Quartz Flexible Accelerometer-based parametric modeling and forecasting NAR dynamic neural network[J]. Journal of Southwest University of Science and Technology, 2016, 31(1): 88-92.
[27]	LI Xiaopeng. Comparing the Kalman filter with a Monte Carlo-based artificial neural network in the INS/GPS vector gravimetric system[J]. Journal of Geodesy, 2009, 83(9): 797-804. doi: 10.1007/s00190-008-0293-y
[28]	CHENG B, TONG H. On consistent nonparametric order determination and chaos[J]. Journal of the Royal Statistical Society: Series B (Methodological), 1992, 54(2): 427-449. doi: 10.1111/rssb.1992.54.issue-2
[29]	LI Wentao, YAN Xiong, XIA Lei, et al. Abnormal data detection and process by using BDS satellite offset semiparametric adjustment model[J]. Acta Geodaetica et Cartographica Sinica, 2020, 49(1): 55-64. DOI: 10.11947/j.AGCS.2020.20180384. doi: 10.11947/j.AGCS.2020.20180384
[30]	ZHANG Yipeng, CHEN Liang, HAO Huan. LM based training algorithm for quantum neural networks[J]. Computer Science, 2013, 40(9): 221-224.

PRN	NAR		W-NAR			WM-NAR
PRN	RMSE	Range		RMSE	Range		RMSE	Range
02	24.973	54.042	21.336		47.432	20.154		45.223
08	4.765	13.431	4.442		10.312	4.115		8.262
10	219.88	608.41	192.47		572.642	175.324		523.521
24	38.557	90.910	30.543		81.342	29.032		78.253
27	62.545	109.684	61.324		94.453	60.012		89.231
31	71.251	20.747	67.123		18.175	53.678		17.423
All	70.328	149.537	62.873		137.393	57.053		126.985

PRN		QP			GM(1,1)				ARIMA				WNN				WM-NAR
PRN		RMSE	Range			RMSE	Range			RMSE	Range			RMSE	Range			RMSE	Range
02		56.521	99.115		169.150		311.350		58.951		114.830		14.321		26.566		20.154		45.223
08		5.783	19.909		79.137		136.640		16.904		28.518		1.111		5.292		4.115		8.262
10		167.04	301.59		1082.5		2046.1		369.57		673.72		187.367		632.581		175.324		523.521
24		10.200	4.057		2.661		4.973		12.273		40.278		34.365		89.171		29.032		78.253
27		68.445	122.800		766.030		1449.3		120.78		222.85		89.421		101.272		60.012		89.231
31		11.357	16.863		575.680		1115.0		9.123		14.451		70.432		26.342		53.678		17.423
ALL		53.224	94.056		445.860		843.894		97.934		182.441		66.1695		146.8707		57.053		126.985
Time/s	0.54			0.76				2.48				10.8				1.49

PRN		QP			GM(1,1)				ARIMA				WNN				WM-NAR
PRN		RMSE	Range			RMSE	Range			RMSE	Range			RMSE	Range			RMSE	Range
02		161.63	328.560		494.99		1014.7		173.350		360.080		23.324		32.566		37.334		26.115
08		10.722	48.056		221.68		439.570		23.000		40.750		5.631		9.392		6.387		13.126
10		469.660	955.410		3253.4		6756.4		886.070		170.630		212.289		642.325		216.123		596.321
24		34.087	100.310		56.247		112.120		37.649		98.823		51.643		106.723		41.112		78.432
27		195.7	404.9		4804.5		2309.4		304.91		605.89		98.421		112.332		66.371		92.231
31		17.72	12.759		1864.2		4019.3		8.846		14.451		91.567		48.576		57.432		23.552
ALL		148.253	308.333		1782.5		2441.9		238.971		215.1		75.956		144.129		70.793		138.296
Time/s	0.66			0.82				2.67				19.6				1.87

PRN		QP			GM(1,1)				ARIMA				WNN				WM-NAR
PRN		RMSE	Range			RMSE	Range			RMSE	Range			RMSE	Range			RMSE	Range
02		571.040	1251.3		1705.4		3697.8		600.630		1314.4		43.123		59.512		50.235		41.721
08		80.744	188.820		737.99		1623.1		29.603		111.520		10.481		14.532		12.337		21.123
10		1552.6	3310.4		11508.1		25249.4		2398.4		4818.9		247.543		752.369		235.223		672.123
24		131.89	323.51		180.98		389.11		134.78		320.39		72.963		134.334		65.371		109.532
27		693.7	1517.2		8218		18066		916.4		1920.1		111.324		123.324		77.773		91.231
31		22.325	51.988		7872.8		18604		14.19		37.297		112.365		126.554		65.432		32.231
ALL		508.717	1107.2		5037.2		11271.6		682.334		1420.4		99.633		201.771		84.395		161.327
Time/s	0.76			0.89				2.87				32.3				2.02