engleski

Predictive Model of Total Electron Content during Moderately Disturbed Geomagnetic Conditions for GNSS Positioning Performance Improvement

Space weather, geomagnetic, and ionospheric conditions have been identified as the single most influential cause of the Global Navigation Satellite System (GNSS) positioning performance degradation and disruption. The random nature of the appearance, development, and intensity of GNSS ionospheric events renders them particularly hard to describe with a single global model. Moderate to massive space weather disturbances remains an unresolved problem for satellite navigation. Here the problem of modelling Total Electron Content (TEC), an outcome of ionospheric conditions that affect GNSS positioning performance, in moderately disturbed geomagnetic field conditions, is addressed based on the experimental observations of the geomagnetic field and their statistical properties. Here the problem of modelling Total Electron Content in moderately disturbed geomagnetic field conditions, as the originator of the ionospheric conditions, is addressed based on the experimental observations of the geomagnetic field and their statistical properties. Three statistical learning-based models are developed, and their performance is assessed according to their ability to predict the resulting TEC. Statistical learning-based TEC prediction models are assessed for their performance, and recommendations for their utilizations are given. The research will continue with the development of a larger database of observations taken at various observation sites and during differently generated moderate geomagnetic events. This database will allow for the development of more robust TEC prediction models using statistical learning methods for self-adaptive mitigation of GNSS ionospheric effects for the core GNSS positioning performance improvement utilized with a wide range of GNSS applications.

Global Navigation Satellite System ; space weather ; geomagnetic conditions ; Total Electron Content (TEC) ; positioning performance ; statistical learning ; prediction model

The FUSION2020 Conference proceedings is indexed in the Scopus and IEEEXplore databases.