Abstract: Objectives: The aim of this study is to obtain a high-precision co-seismic deformation field of a 6.2 magnitude earthquake occurred in Jishishan County, Gansu Province, China from high frequency GNSS observation data, which will enable the real-time monitoring of deformations for earthquakes with a magnitude of 6 or higher using a massive network of continuously operating reference stations (CORS). Methods: This study employed the Pride 3.0 software developed by Wuhan University for high-precision dynamic processing of 32 CORS station with high- frequency data, obtaining co-seismic deformation time series for each station. For stations showing significant co-seismic responses in close proximity, dynamic solutions with large changes at the seismic moment were first excluded. The smoothed single-epoch results before and after the earthquake were retained. For stations with less obvious co-seismic responses, the Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) method was used to detect high-frequency co-seismic signals. Results: The main range of permanent deformation caused by earthquakes is in the area 30km away from the epicenter. Stations 5km away from the epicenter recorded permanent deformation in the east-west direction of 13mm, north-south direction of 10mm, and vertical deformation of 8mm. Modal decomposition identified 13 stations with seismic responses in the east or north directions, including CORS stations LXJS, GUTI, LXYJ within a 40 km radius from the epicenter. It could detect the seismic deformation far from 100km better than the original GNSS results, which means this method could be used for waveform signal reconstruction, noise removal, and expanding the distance of CORS station monitoring for co-seismic signals. Conclusions: This study indicate that various types of CORS station equipment with different foundation can monitor earthquake signals. The CEEMDAN method for signal decomposition can detect co-seismic signals faraway, providing feasibility for monitoring seismic co-seismic deformation information using a large number of CORS station resources in the future.