GNSS Geodetic Velocity Prediction Using Deep Neural Network: A Case Study of Cairo Area, Egypt

Document Type : Regular Articles

Authors

1 Dept. of Geodynamics, National Research Institute of Astronomy and Geophysics (NRIAG), Helwan, Egypt.

2 Department of Computer Science and Mathematics, Faculty of Science, Aswan University

3 Department of Mathematics, Faculty of Science, Sohag University, Sohag 82511, Egypt.

Abstract

Abstract: Estimating and studying global navigation satellite system (GNSS) velocities play an essential role in understanding the deformation and motion of the crust. Thus, in this research, we employ a deep neural network (DNN) to estimate horizontal velocities at certain places using GNSS data. Data on crustal deformation are obtained by using Global Positioning System (GPS) techniques. The exact locations of the three stations were obtained by recording, analyzing, and adjusting permanent GPS measurements. Moreover, 70% of the GNSS velocities from stations in the Cairo region and International GNSS Service (IGS) stations were used in the analysis to train the proposed DNN model, with the remaining 30% set aside for testing. The horizontal velocity components (east and north) were estimated using the DNN model. The highest differences between the velocities obtained by the DNN model and the reference velocities were 0.0004 mm. These findings highlight the ability of the DNN model to provide precise GNSS velocity estimates for geodetic applications.

Keywords

Main Subjects

Files

Share

How to cite

Statistics