Abstract
During permanent GNSS observations a displacement of the network’s station may occur. Not all changes are significant enough to raise an alarm. However, the critical value of the station’s position change may vary significantly for various reasons, such as the antenna model, receiver or observation interval. The coordinates gained by GNSS methods are also usually characterized by a higher degree of precision of the horizontal components than those of the vertical ones. It is virtually impossible to set a single critical value for each component of each station of the network; therefore, an automatic algorithm evaluating the estimates should be used. The paper suggests an adaptive algorithm using a combination of the Kalman filter and a modified moving average technique to reveal sudden changes in a GNSS station’s position, while allowing it to adapt to long-term changes.