Is this project an undergraduate, graduate, or faculty project?
Undergraduate
Loading...
group
What campus are you from?
Daytona Beach
Authors' Class Standing
Jose Nicolas Gachancipa, Senior Chintan Thakrar, Junior
Lead Presenter's Name
Jose Nicolas Gachancipa
Faculty Mentor Name
Kshitija Deshpande
Abstract
The ionosphere is a region in the Earth’s upper atmosphere, where atoms are ionized due to solar radiation. The behavior of the ionosphere depends on time and location, and it is highly influenced by solar activity. The ionization process creates layers of free electrons at different altitudes, which can cause fluctuations in electromagnetic waves crossing the region. The effect of ionospheric events on radio signals can be measured using Global Navigation Satellite Systems (GNSS) receivers, in terms of ionospheric scintillation and Total Electron Content (TEC). The GNSS team at the Space Physics Research Lab (SPRL) studies ionospheric events using multi-frequency GNSS receivers (NovAtel GPStation-6) capable measuring high and low rate scintillation data as well as TEC values from three different GNSS systems (GPS, GALILEO, and GLONASS).
The purpose of this project is to develop a machine learning algorithm, using recurrent neural networks, to detect ionospheric events in low-rate scintillation data. Recurrent neural networks are often used for time-series applications, including forecasting and prediction. The model is being trained using data collected by the GNSS receivers in multiple locations (including Daytona Beach), with a focus on high-latitude data from the Canadian High Artic Ionospheric Network (CHAIN). The machine learning model will be integrated with the Embry-Riddle Ionospheric Scintillation Algorithm (EISA), an existing model capable of processing ionospheric data. EISA was developed by the GNSS team at SPRL. The updated model will allow the team to automate the process of ionospheric event detection, which is currently done manually. Upon this implementation, EISA will become an end-to-end model for ionospheric data collection, processing, and modelling.
Did this research project receive funding support from the Office of Undergraduate Research.
Yes, Ignite Grant
Implementation of Machine Learning Methods for Ionospheric Scintillation Data Analysis
The ionosphere is a region in the Earth’s upper atmosphere, where atoms are ionized due to solar radiation. The behavior of the ionosphere depends on time and location, and it is highly influenced by solar activity. The ionization process creates layers of free electrons at different altitudes, which can cause fluctuations in electromagnetic waves crossing the region. The effect of ionospheric events on radio signals can be measured using Global Navigation Satellite Systems (GNSS) receivers, in terms of ionospheric scintillation and Total Electron Content (TEC). The GNSS team at the Space Physics Research Lab (SPRL) studies ionospheric events using multi-frequency GNSS receivers (NovAtel GPStation-6) capable measuring high and low rate scintillation data as well as TEC values from three different GNSS systems (GPS, GALILEO, and GLONASS).
The purpose of this project is to develop a machine learning algorithm, using recurrent neural networks, to detect ionospheric events in low-rate scintillation data. Recurrent neural networks are often used for time-series applications, including forecasting and prediction. The model is being trained using data collected by the GNSS receivers in multiple locations (including Daytona Beach), with a focus on high-latitude data from the Canadian High Artic Ionospheric Network (CHAIN). The machine learning model will be integrated with the Embry-Riddle Ionospheric Scintillation Algorithm (EISA), an existing model capable of processing ionospheric data. EISA was developed by the GNSS team at SPRL. The updated model will allow the team to automate the process of ionospheric event detection, which is currently done manually. Upon this implementation, EISA will become an end-to-end model for ionospheric data collection, processing, and modelling.