Exploration of machine learning tools developed for the study of space weather and its impact on position approximation in GNSS systems
Reporte
9 de julio de 2021
Resumen:
The equatorial ionosphere has been extensively studied using purely physical models, however in recent years, with a large amount of data, it has been possible to improve these models using machine learning techniques. In this paper, we share the research results aimed to evaluate the influence of space weather parameters on GPS position approximation. We evaluated data from the Huancayo GPS station between 2016 and 2020 and we have taken into account the space weather data from the OMNI website, scintillation index (S4) and position data obtained from the GPS of the LISN network to perform our model. In addition, we use tropospheric conditions provided by the Geophysical Institute of Peru (IGP). The final result is a reliability matrix obtained with an XG Boost algorithm that will allow us to evaluate if a GPS signal given the conditions is indeed reliable or not.
Autores:
G. Fajardo; E. Pacheco
Fecha:
2021-06
URL:
http://hdl.handle.net/20.500.12816/4963
The equatorial ionosphere has been extensively studied using purely physical models, however in recent years, with a large amount of data, it has been possible to improve these models using machine learning techniques. In this paper, we share the research results aimed to evaluate the influence of space weather parameters on GPS position approximation. We evaluated data from the Huancayo GPS station between 2016 and 2020 and we have taken into account the space weather data from the OMNI website, scintillation index (S4) and position data obtained from the GPS of the LISN network to perform our model. In addition, we use tropospheric conditions provided by the Geophysical Institute of Peru (IGP). The final result is a reliability matrix obtained with an XG Boost algorithm that will allow us to evaluate if a GPS signal given the conditions is indeed reliable or not.
Autores:
G. Fajardo; E. Pacheco
Fecha:
2021-06
URL:
http://hdl.handle.net/20.500.12816/4963
Esta publicación pertenece al compendio Repositorio Geofísico Nacional (REGEN)
Repositorio Geofisico Nacional.pdf
PDF
256.8 KB