Fuzzy weighting approach for single point positioning with single-frequency pseudo-range observations

Abstract

The Single-frequency receiver based on pseudo-range single point positioning (SPP) technology is widely used due to its advantage of low cost. As the weighted least squares (WLS) estimation in SPP is used to calculate the user’s position, optimizing weight estimation may be a strategy to improve the positioning accuracy. So far, lots of effective weighting approaches have been developed, such as satellites elevation angle, signal-to-noise ratio (SNR), measurement residual and priori-variances weighting approach. Most of these weighting approaches use one factor or more factors without considering the correlation between them to estimate the weight matrix, possibly leading to imprecise weight estimation. However, the situation in building weighting model with multiple factors and their correlation are complicated and difficult. In this study, we try to develop a new weighting approach called fuzzy weighting approach to improve the positioning accuracy. This approach adopts the fuzzy control theory to estimate the weights of the observations. It does not limit numbers of factors involving into the weight estimation and require the precise mathematic model describing the relation between factors and weight. This study presents the results of the static and kinematic experiments of pseudo-range SPP with WLS using four weighting approaches: SNR-based, elevation-dependent, combined SNR and elevation and fuzzy weighting approaches. The results show fuzzy weighting approach more effectively improves the positioning accuracy.