Abstract: Complex monitoring environments and deformation modes can lead to Global Navigation Satellite System (GNSS) time series in deformation monitoring containing multi-source heterogeneous errors and signals. To meet the requirements for high-precision deformation signal extraction from GNSS time series, this study proposed an automated processing method for GNSS time series based on the separate correction of outliers, step terms, and error values. First, the numerical variation characteristics of various error terms are discussed and analyzed. Next, correction strategies-utilizing mean and standard deviation, data continuity checks, and deformation characteristic function fitting-are applied to correct outliers, step terms, and error values in the GNSS time series. This approach enhances the accuracy and reliability of the GNSS time series to obtain precise deformation signals. Finally, the method is validated using nearly eight years of observational data from 43 GNSS stations on Sumatra Island. The results show that the displacement rates obtained after processing exhibit similar variation characteristics consistent with actual conditions: eastward movement of 20~35 mm/a in the E direction, northward movement of 15~30 mm/a in the N direction, and uplift or subsidence rates within 10 mm/a in the U direction. This indicates that the method is reliable and effective for rate estimation. Experiments demonstrate that the proposed method can systematically perform outlier removal, step term correction, deformation characteristics parameter estimation, and uncertainty evaluation, making it suitable for the automated processing of GNSS time series in complex environments such as coastal areas.