This paper used improved Levenberg-Marquart BP neural network technology to predict GPS tropospheric delay with the data of Beijing GPS data and meteorological data, which the input parameters are either space position (longitude, latitude, elevation) or space position, temperature, pressure. The testing results showed that the two predicted results are mostly closer; meanwhile the prediction results considering space position, temperature and pressure are slightly superior to the other method. The deviations between the predicted values of most GPS Stations and the actual atmospheric delay values are about 5mm. The accuracy of tropospheric delay predicted by BP neural network amounted to millimeters. © 2012 Springer-Verlag GmbH.
CITATION STYLE
Wang, Y., Zhang, L., & Yang, J. (2012). Prediction of Zenith tropospheric delay based on BP neural network. In Advances in Intelligent and Soft Computing (Vol. 140 AISC, pp. 459–465). https://doi.org/10.1007/978-3-642-27945-4_73
Mendeley helps you to discover research relevant for your work.