Abstract: In this paper, the wavelet analysis method was applied to the data processing of fixed-point deformation observation data. Frequency anomalies were extracted through the spatiotemporal scanning of water pipe inclination and cave strain data at 6 stations with good continuity of observation data and reliable data in Yunnan Province from 2008 to 2018. Statistical analysis of the spatiotemporal correspondence between the anomalies and M_S≥5 earthquakes in Yunnan area was put forward, the purpose of which is to obtain some reference value of wavelet analysis in daily forecasting. The research results showed that the reflecting effect of multiple synchronous anomalies extracted by wavelet analysis was fairly good, and within 6 months after the occurrence of the anomaly, the earthquake occurrence rate of M_S≥5 earthquakes in Yunnan area reached 71%. Furthermore, frequency synchronization anomalies could be observed before the M_S≥6 earthquakes. Meanwhile, it was found out that the anomalies extracted by wavelet analysis could be used in daily tracking to determine the time and magnitude of future earthquakes.