GPS Error Modeling Based on Time Series and Research on the Accuracy of Point Positioning
Zou Wei-jun · Acta Armamentarii · 2009
A data processing algorithm was proposed to improve the GPS single point positioning accuracy.Full details were provided on how to use time series analysis in the modeling of the first order differential series for the GPS error.The AR(n) model which is adopted to describe the first order differential series for the GPS error was derived from the observed data,and model parameter values were obtained using the least square method;one step prediction for the error series was effectively achieved by conditional expectation;positioning accuracy of ordinary GPS receivers was improved remarkably through online adjustment to the observation data.The simulated results show efficiency and reliability of the proposed method.