eqing Xu

Contributions of thermoelastic deformation to seasonal variations in GPS station position

We investigate surface displacements due to land temperature variation with the 2014 global thermoelastic model, which is a solution on a uniformly elastic sphere under the constraint that the geocenter remains stationary. In this research, the seasonal variations of global surface displacements are numerically simulated based on 0–10xa0cm underground land surface temperatures from National Oceanic and Atmospheric Administration. The displacements include vertical and horizontal components for the first time. Meanwhile, the annual contributions of geophysical sources, which are mainly due to atmosphere, ocean, snow and continental water, are also estimated. For comparative analyses, the partial displacement by annual mass-loading and the total displacement by the combined annual of thermoelasticity and mass-loading are calculated, respectively, and displayed against the annual displacements at stations of global positioning system network. Results of the numerical simulation show that the amplitude of surface thermoelastic deformation is at the millimeter level on the global scale, topped at about 3xa0mm for radial displacement and about 1.5xa0mm for transverse components, which need to be considered for the high-precision terrestrial reference frame. The combined deformation caused by thermoelastic and mass-loading can explain the seasonal GPS observations better than the mass-loading alone, in particular for the transverse displacements.

Combined Prediction of Earth Orientation Parameters

Earth orientation parameters (EOP) are essential for transformation between the celestial and terrestrial coordinate systems, which has important applications in the Earth sciences, astronomy and satellite navigation. The latter cannot be considered as self-reliant without accurate EOP predictions, which, in particular, are required for real-time precise orbit determination. In this paper we firstly describe the principles and analyze the characteristics of several EOP prediction methods. Then, according to the forecast accuracy, the weights are assigned to EOP predictions to get a combined solution. Results show that no single forecasting method is suitable for all the parameters at different time spans. The combined solution integrates advantages of different prediction methods, which avoid the limitations and instability of any of them. The resultant combined prediction is issued to the JPL Earth Orientation Parameters Combination of Prediction Pilot Project (EOPC PPP).

Application of the radial basis function neural network to the short term prediction of the Earth’s polar motion

By a number of test cases using different sample numbers and sample lengths, we obtain a Radial Basis Function Neural Network (RBFNN) model that is suitable for the short-term forecast of polar motion, especially for the ultra-short-term forecast. By using the same data sample of Earth’s polar motion, this RBFNN model can achieve better short-term prediction accuracy than the least-squares+autoregressive (LS+AR) method, and better ultra-short-term prediction accuracy than the LS+AR+Kalman method. Using this model to forecast the polar motion data from January 1, 2002 to December 30, 2007 and from January 1, 2010 to December 30, 2016, respectively, experimental results show that the ultra-short-term forecast accuracy of this RBFNN model is within a precision of 3.15 and 3.08 milliseconds of arc (mas) in polar motion x direction, 2.02 and 2.04 mas in polar motion y direction; the short-term forecast accuracy of RBFNN model is within a precision of 8.83 and 8.69 mas in polar motion x direction, and 5.59 and 5.85 mas in polar motion y direction. As is stated above, this RBFNN model is well capable of forecasting the short-term of polar motion, especially the ultra-short-term.

Research on High Accuracy Prediction Model of Satellite Clock Bias

Time basis of satellite navigation system is achieved by the satellite clock bias (SCB) prediction, while the SCB prediction accuracy will also affect the positioning accuracy of real-time navigation users. With the development of our Beidou satellite navigation system, the accuracy requirements of the SCB prediction is higher and higher, general quadratic polynomial extrapolation method have failed to meet the SCB forecast accuracy for each satellite. There is an urgent need to develop the SCB prediction program for each satellite, here we use a combined method of least squares and auto-regressive model (LS + AR) from the EOP forecasting, to predict and assess SCB with data from IGS. Results show that the combined LS + AR method can improve the SCB forecast accuracy effectively.

Performance Evaluation of the Beidou Satellite Clock and Prediction Analysis of Satellite Clock Bias

Satellite clock bias (SCB) is provided by the in orbit atomic clock, which is the key to satellite navigation system. First, the time reference of the satellite navigation system is realized by the SCB, and the SCB prediction accuracy will also affect the positioning accuracy of real-time navigation users. With the development of our Beidou satellite navigation system (BDS), the performance requirement of the satellite clock and accuracy requirement of the SCB prediction are higher and higher. This paper will study on performance evaluation of BDS atomic clock and the prediction analysis of SCB series exclusively. In order to show the results objectively and effectively, we select two data processing centers of the GeoForschungsZentrum Potsdam (GBM), and SHAO Analysis Center (SHA), to obtain the same time BDS clock bias sequence as the base data, for a comparative analysis. First, the performance of the atomic clock is evaluated by statistics of the Allen variance. Meanwhile, we establish the model for each SCB sequence according to the characteristic of the atomic clock, by using a combined method of least squares and autoregressive model (LS+AR), to predict and assess the SCB with root mean square error (RMS). Results show that the performance of BDS in orbit atomic clock is stable, with the day stability in the order of 10−14; And the atomic clock performance is related to the SCB prediction accuracy, that is shown as better performance with higher prediction accuracy; Mean while the LS+AR model predict SCB series based on the performance of different atomic clocks, which can improve the SCB prediction accuracy effectively.