Multipath Mitigation for Single Frequency Stand-Alone Receivers Using Wavelet Denoising

Abstract

Multipath error is one of the major error sources impacting high precision positioning in Global Navigation Satellite Systems (GNSS). Particularly, for commonly available single frequency (SF) receivers, an unaccounted multipath error can severely affect positioning accuracy. For multipath mitigation, SF code measurements can be filtered and these filtered pseudoranges are used for positioning. However, filtering multipath is a challenging problem due its stochastic nature. Wavelet signal processing, used for analysis of non-stationary data, can cater for the stochastic nature of multipath. This paper uses wavelet denoising scheme for reducing multipath error in SF code-pseudoranges. The procedure for selecting wavelet filters is outlined. Finally, the positions obtained with wavelet filtered pseudoranges are compared to positions computed with unfiltered pseudoranges. It is seen that wavelet denoising refines the GNSS observation data by reducing multipath and gives smoothed pseudoranges resulting in higher positioning precision. Therefore, an appropriate selection of wavelet filters results in reduction of multipath to a great extent for SF measurements.