A noise analysis method for GNSS signals of a standalone receiver

Extensive studies on signal noise analysis of Global Navigation Satellite Systems (GNSS) receivers are mainly based on zero and/or short baselines where a pair of receivers is required. This paper develops a new signal noise assessment method for a standalone GNSS receiver. In this new method, the time differenced geometry-free (Δ\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ \Delta $$\end{document}GF) model is formed, where the residual ionospheric delay and multipath remain. In order to eliminate these systematic trend biases, the second-order polynomial fitting is used. Then the mixed autoregressive moving average (ARMA) model is further applied to capture this stationary and time-correlated time series. As a result, the pure random noise of Δ\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ \Delta $$\end{document}GF is obtained, which is finally used to assess the precision of receiver signal in terms of the error propagation law. The performance of this new method is numerically tested by using 10 sets of data and compared with the methods of zero and short baselines. The results indicate that the new method is able to assess both the phase and code precision of a standalone receiver, and the conclusions are consistent to those from the methods of zero and short baselines. Since again the new method needs only one receiver, it could be implemented in more applications than the traditional methods, especially in precise point positioning.