基于相空间重构与高斯过程的卫星钟差预报

提出了一种基于相空间重构与高斯过程预报卫星钟差的新方法。首先根据星载原子钟的物理特性用多项式进行拟合以提取钟差趋势项,并对拟合后的残差进行经验模态分解,作降噪处理;然后以降噪后的残差时间序列的混沌特性为基础,对其进行相空间重构;最后以重构的相空间为基础,运用高斯过程对残差时间序列进行建模预报,再将预报结果加上趋势项,获得最终的钟差预报值。采用IGS提供的GPS超快速观测钟差建模进行短期预报实验,结果表明,该方法能实时有效地对卫星钟差进行预报,且精度优于超快速预报钟差。

A Novel Method for Satellite CloCk Bias PrediCtion Based on Phase SpaCe ReConstruCtion and Gaussian ProCesses

A novel method for satellite clock bias prediction incorporating phase space reconstruction and Gaussian processes(GP)is proposed. Firstly a polynomial model is employed to fit the time-series of clock bias in terms of its characteristics,and then the noise of the polynomial fitting residuals is reduced by the empirical mode decomposition (EMD)algorithm. Secondly,phase space reconstruction is performed for the de-noised residuas according to its chaotic characteristics. Finally,Gaussian processes are established so as to forecast the residuals on the basis of reconstructed phase space,and then the predicted clock bias can be yielded by adding the trend term to the forecasted residuals. The IGS ultra-rapid observed(IGU-O)product is used to set up a clock model,and short-term prediction experiments are carried out. The results show that the proposed method outperforms the IGS ultra-rapid predicted(IGU-P)solutions at least on a daily basis.