Wavelet Transform Based Kalman Filtering Algorithm for Anti-SA Effect

Abstract

A new mathematical approach, wavelet transformation, is introduced for GPS data analysis. Based on its good features on both time-domain and frequency-domain, one can obtain a true time-frequency representation of a signal. The so-called mutiresolution analysis (MRA) will be used for data analysis, such as for the removal of noises, and the detection and rejection of gross errors from signal. In this paper the authors present a wavelet-based algorithm for the modeling and prediction of SA effect by combining time series analysis methods. A new self-adaptive Kalman filtering algorithm is presented for anti-SA. Some preliminary test results from experimental data are also summarized.