Identification and Processing of Abnormal Data in Differential GNSS Dynamic Measurement

In order to suppress the effect of abnormal data in differential GNSS dynamic measurement, an identification and processing method based on first-order differential time series and sliding window plus bidirectional fitting extrapolation is proposed. The method firstly identifies and locates the lost data based on time series, then completes the forward and reverse fitting extrapolation of the outliers by sliding window of fixed length point by point, and obtains the weighted fusion of estimation of the outliers according to the inverse distance weighting method. The simulation results show that the method can solve the identification and extrapolation problem of lost data points and outliers in the whole curve, and does not affect the normal data.