Anna-Marie Bals
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Anna-Marie Bals has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Kshitija Bharat Deshpande
Department of Physical Sciences and Center for Space and Atmospheric Research (CSAR) Embry‐Riddle Aeronautical University Daytona Beach FL USA
2 shared publications - Larry J. Paxton
The Johns Hopkins University Applied Physics Laboratory Laurel MD USA
2 shared publications Chintan Thakrar
Department of Physical Sciences and Center for Space and Atmospheric Research, Embry-Riddle Aeronautical University, Daytona Beach, FL, USA
1 shared publication- Chintan Thakrar
Center for Space and Atmospheric Research (CSAR) Embry‐Riddle Aeronautical University Daytona Beach FL USA
1 shared publication Kshitija B. Deshpande
Department of Physical Sciences and Center for Space and Atmospheric Research, Embry-Riddle Aeronautical University, Daytona Beach, FL, USA
1 shared publicationNicolas Gachancipa
Center for Space and Atmospheric Research (CSAR) Embry‐Riddle Aeronautical University Daytona Beach FL USA
1 shared publication
Latest Publications
- Open Access
- 17 February 2026
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Machine Learning and Computation
- Vol. 3 (1)
Categorizing Characteristic Regions of High‐Latitude Scintillations Using a Combination of Isolation Forest and Neural Network Machine Learning Algorithms
Global Positioning System (GPS) scintillations are radio signal fluctuations due to ionospheric structures or irregularities. In this work, we utilize scintillation data to identify the source region (auroral oval vs. polar cap) of a high latitude irregularity responsible for any given scintillation instance, using a combination of ML and deep learning models. We used available high‐rate (50‐Hz) GPS data from three Canadian High Arctic Ionospheric Network stations for 2 yrs to develop a categorization methodology based on an unsupervised detection ML model called Isolation Forest (iForest), which detects scintillation instances to be fed to a supervised artificial neural network (ANN) model. The goal is to confidently categorize the high‐latitude ionospheric scintillations based on their regions of occurrence. Our method involves using low‐rate GPS scintillation indices to threshold high‐rate data to be fed to the iForest for event detection. We use satellite data‐derived auroral oval and polar cap boundaries to label our ANN data set. Our analysis of these preliminary data sets shows that the iForest algorithm detects events with high accuracy irrespective of the geomagnetic conditions and receiver location. The ANN consistently yielded an F1‐score close to 0.7 implying that the model can classify source regions based on the input data. The ANN‐based classification of source regions of scintillation events performed better during quiet times than stronger geomagnetic conditions. Our method can be applied to different ionospheric irregularity problems, making this a potentially useful tool for understanding scintillations and their relationship with different irregularity generation mechanisms.
- 17 February 2026
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Machine Learning and Computation
- Vol. 3 (1)
Open AccessCategorizing Characteristic Regions of High‐Latitude Scintillations Using a Combination of Isolation Forest and Neural Network Machine Learning Algorithms
Global Positioning System (GPS) scintillations are radio signal fluctuations due to ionospheric structures or irregularities. In this work, we utilize scintillation data to identify the source region (auroral oval vs. polar cap) of a high latitude irregularity responsible for any given scintillation instance, using a combination of ML and deep learning models. We used available high‐rate (50‐Hz) GPS data from three Canadian High Arctic Ionospheric Network stations for 2 yrs to develop a categorization methodology based on an unsupervised detection ML model called Isolation Forest (iForest), which detects scintillation instances to be fed to a supervised artificial neural network (ANN) model. The goal is to confidently categorize the high‐latitude ionospheric scintillations based on their regions of occurrence. Our method involves using low‐rate GPS scintillation indices to threshold high‐rate data to be fed to the iForest for event detection. We use satellite data‐derived auroral oval and polar cap boundaries to label our ANN data set. Our analysis of these preliminary data sets shows that the iForest algorithm detects events with high accuracy irrespective of the geomagnetic conditions and receiver location. The ANN consistently yielded an F1‐score close to 0.7 implying that the model can classify source regions based on the input data. The ANN‐based classification of source regions of scintillation events performed better during quiet times than stronger geomagnetic conditions. Our method can be applied to different ionospheric irregularity problems, making this a potentially useful tool for understanding scintillations and their relationship with different irregularity generation mechanisms.
- Open Access
- 18 September 2025
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Machine Learning and Computation
- Vol. 2 (3)
Classification of High‐Latitude Source Region‐Based Ionospheric Scintillation Signatures Using Machine Learning
This study investigates ionospheric scintillation signatures in GNSS signals at high latitudes with Machine Learning (ML), focusing on different irregularity generation mechanisms in major regions of the ionosphere. The diverse structuring in various scale sizes resulting from these mechanisms might produce distinct scintillation signatures. In the nightside ionosphere, precipitation is predominantly observed in the auroral oval versus density gradients in the polar cap. We apply this physics‐based precursor knowledge about the source regions and establish a database of scintillation events from significant geomagnetic activity to explore these differences in scintillation signatures. Scintillation event segments in high‐rate (50 Hz) Global Navigation Satellite System (GNSS) signal phase and amplitude are selected as input data to analyze time‐dependent signature features. These pre‐processed events are collected from stations in the auroral oval and polar cap regions and labeled based on ionospheric conditions during four geomagnetic storms. Using this database, we train an ML model to classify events into polar cap versus auroral oval signatures in two systematic steps. Initially, we compare signatures from two stations in different regions with hierarchical clustering, which applies correlation analysis to group them based on the similarity of their time series features. We train a decision tree model with the correlation results to classify the clustered time series according to their predicted location (auroral vs. polar). We present an ML‐focused methodology connecting two predominant high‐latitude ionospheric regions with their distinctly observed families of scintillation signatures. This research helps trace the irregularity mechanisms present in these regions through GNSS observations.
- 18 September 2025
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Machine Learning and Computation
- Vol. 2 (3)
Open AccessClassification of High‐Latitude Source Region‐Based Ionospheric Scintillation Signatures Using Machine Learning
This study investigates ionospheric scintillation signatures in GNSS signals at high latitudes with Machine Learning (ML), focusing on different irregularity generation mechanisms in major regions of the ionosphere. The diverse structuring in various scale sizes resulting from these mechanisms might produce distinct scintillation signatures. In the nightside ionosphere, precipitation is predominantly observed in the auroral oval versus density gradients in the polar cap. We apply this physics‐based precursor knowledge about the source regions and establish a database of scintillation events from significant geomagnetic activity to explore these differences in scintillation signatures. Scintillation event segments in high‐rate (50 Hz) Global Navigation Satellite System (GNSS) signal phase and amplitude are selected as input data to analyze time‐dependent signature features. These pre‐processed events are collected from stations in the auroral oval and polar cap regions and labeled based on ionospheric conditions during four geomagnetic storms. Using this database, we train an ML model to classify events into polar cap versus auroral oval signatures in two systematic steps. Initially, we compare signatures from two stations in different regions with hierarchical clustering, which applies correlation analysis to group them based on the similarity of their time series features. We train a decision tree model with the correlation results to classify the clustered time series according to their predicted location (auroral vs. polar). We present an ML‐focused methodology connecting two predominant high‐latitude ionospheric regions with their distinctly observed families of scintillation signatures. This research helps trace the irregularity mechanisms present in these regions through GNSS observations.
- 7 April 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Radio Frequency Identification
- Vol. 6, 240-249
Creating a Database to Identify High-Latitude Scintillation Signatures With Unsupervised Machine Learning
In high latitudes, Global Navigation Satellite System (GNSS) signals experience scintillation due to moving irregularity structures in the ionosphere. These develop as a result of different physical mechanisms, which are as yet principally described on an elementary level for certain storm cases and events. Since there are years of GNSS data available from stations around the globe, we are investigating an unsupervised Machine Learning approach to extract a large variety of groups of scintillation events with similar features. We create a database containing high-rate scintillation events from two geomagnetic storm cases and several stations in the high-latitude region of the Northern hemisphere. By clustering high-rate signatures in signal phase and power according to their major signal characteristics with an agglomerative hierarchical clustering, it is possible to extract different groups of similar types of scintillation signatures. As a result of this study, the database of scintillation signatures in various locations in the auroral oval and polar cap evolves and will be further expanded beyond the storm cases studied in this paper. These can then be linked to the geomagnetic conditions and dynamics in the ionosphere through additional datasets from other instruments, therefore potentially helping us to get a further insight into the ionospheric irregularity physics.
- 7 April 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Radio Frequency Identification
- Vol. 6, 240-249
Creating a Database to Identify High-Latitude Scintillation Signatures With Unsupervised Machine Learning
In high latitudes, Global Navigation Satellite System (GNSS) signals experience scintillation due to moving irregularity structures in the ionosphere. These develop as a result of different physical mechanisms, which are as yet principally described on an elementary level for certain storm cases and events. Since there are years of GNSS data available from stations around the globe, we are investigating an unsupervised Machine Learning approach to extract a large variety of groups of scintillation events with similar features. We create a database containing high-rate scintillation events from two geomagnetic storm cases and several stations in the high-latitude region of the Northern hemisphere. By clustering high-rate signatures in signal phase and power according to their major signal characteristics with an agglomerative hierarchical clustering, it is possible to extract different groups of similar types of scintillation signatures. As a result of this study, the database of scintillation signatures in various locations in the auroral oval and polar cap evolves and will be further expanded beyond the storm cases studied in this paper. These can then be linked to the geomagnetic conditions and dynamics in the ionosphere through additional datasets from other instruments, therefore potentially helping us to get a further insight into the ionospheric irregularity physics.