A GNSS Time Series Denoising Method with Mixed Use of Cross-Validation and CEEMD-WT

Objectives:  The coordinate time series derived from Global navigation satellite system (GNSS) technology usually contains noise, which may lead to misinterpretation of some geophysical phenomena. To reduce to noise level of GNSS datasets, a denoising method based on the mixed use of cross-validation (CV), complementary ensemble empirical mode decomposition (CEEMD) and wavelet transform (WT) is proposed.   Methods:  Specifically, the original GNSS time series is first decomposed into several Intrinsic Mode Function (IMF) components by CEEMD. Then, the CV strategy is applied to divide the IMFs into signal with slight noise components and pure noise components. Finally, after removing the pure noise components, wavelet denoising method is adopted to eliminate the noise in the remaining IMFs to obtain final GNSS time series.   Results:  Both simulated signals and real-world datasets collected from 117 GNSS stations were used to evaluate the performance of the proposed method. Results show that the proposed method can effectively attenuate the noise in the original time series. Compared with the residuals of the original time series, the residuals of denoising results derived from the proposed method are reduced by 43%, 43%, and 46% in the north (N), east (E), and up (U) directions, respectively.   Conclusions:  Compared with wavelet transform, CV-based CEEMD and correlation coefficient-based CEEMD methods, the proposed method can effectively reduce the noise effects in GNSS time series.