Pseudo-range Observations Ionospheric Error Separation Based on BP Neural Network

Yanglin LI,Wende HUANG,Liyuan SHENG

DOI: https://doi.org/10.13442/j.gnss.1008-9268.2015.06.001

2015-01-01

Abstract:Aiming at the problem of ionospheric delay error ,proposed a new error sepa‐ration method based on BP neural network .The training sample is based on a week of obser‐vation data and set the pseudo‐distance ionospheric delay error separation‐related factors‐time ,satellite elevation angle ,azimuth and pseudo‐range observations as BP neural network input unit ,use CODE published high‐precision GIM calculated corresponding ionospheric de‐lay error as an output unit for BP neural network training .Finally ,use the trained model for each satellite to separate next week ionospheric delay error of pseudorange observations .The results show that :the use of BP neural netw ork model to separate ionospheric delay error of pseudo‐range observations ,both of the separation accurate is above 80% ,and the best is up to 90% .The method used a large number of actual historical data ,it is a good attempt to solve navigation problems from the perspective of big data .