Robust Edge Machine-Learning For The Real-Time Processing And Prediction of Geomagnetic Anomalies and Geomagnetically Induced Currents (GICs)

Aug 4, 2026

Upload by Eden

Doctoral dissertation on robust edge machine-learning methods for real-time processing and prediction of geomagnetic anomalies and geomagnetically induced currents (GICs). Motivates the need for accurate space-weather-related forecasting and monitoring, contrasts physics-based simulations with data-driven machine learning, and addresses challenges in event modeling and observation using magnetometers and baseline correction. Establishes research objectives and organization of the work, supported by academic committees, funding from NSF EPSCOR, and acknowledgements to collaborators and lab members.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages139

Reading Time5 h 50 min

AccessFree to download

Introduction

Motivation, Objective, and Research Overview

Research Impact

Organization of Dissertation

Geomagnetic Event Forecasting and Monitoring: Background, Challenges and Research Objective

Introduction

Overview of Geomagnetism and Geomagnetic Events

Approaches to Geomagnetic Event Forecasting: Physics-Based Simulations vs. Machine Learning (ML)-Driven Methods

Approaches to Geomagnetic Event Monitoring: Magnetometers and Baseline Correction

University of New Hampshire University of New Hampshire Scholars Repository | Doctoral Dissertations | Graduate and Undergraduate Student Scholarship | | --- | --- | | Spring 2025
Robust Edge Machine-Learning For The Real-Time Processing And Prediction of Geomagnetic Anomalies and Geomagnetically Induced Currents (GICs)
Talha Siddique
University of New Hampshire
Follow this and additional works at: [https://scholars.unh.edu/dissertation](https://scholars.unh.edu/dissertation) | | Recommended Citation Siddique, Talha, "Robust Edge Machine-Learning For The Real-Time Processing And Prediction of Geomagnetic Anomalies and Geomagnetically Induced Currents (GICs)" (2025) . Doctoral Dissertations. 2881. [https://scholars.unh.edu/dissertation/2881](https://scholars.unh.edu/dissertation/2881) This Dissertation is brought to you for free and open access by the Graduate and Undergraduate Student Scholarship at University of New Hampshire Scholars Repository. It has been accepted for inclusion in Doctoral Dissertations by an authorized administrator of University of New Hampshire Scholars Repository. For more information, [please contact Scholarly.Communication@unh.edu](please contact Scholarly.Communication@unh.edu). Robust Edge Machine-Learning For The Real-Time Processing And Prediction of Geomagnetic Anomalies and Geomagnetically Induced Currents (GICs) BY TALHA SIDDIQUE BS Computer Science, BRAC University, 2016 MS Natural Resources: Environmental Economics, University of New Hampshire, 2019 PhD Dissertation Submitted to the University of New Hampshire in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in Electrical and Computer Engineering May, 2025 This dissertation will be examined and approved in partial fulfillment of the requirements for the degree of PhD in Electrical and Computer Engineering by: Thesis Director, Md Shaad Mahmud, PhD. Associate Professor Department of Electrical and Computer Engineering University of New Hampshire Se Young Yoon, PhD. Associate Professor Department of Electrical and Computer Engineering University of New Hampshire Diliang Chen, PhD. Assistant Professor Department of Electrical and Computer Engineering University of New Hampshire Donpeng Xu, PhD. Associate Professor Department of Computer Science University of New Hampshire Hyunju K Connor, PhD. Research Astrophysicist NASA Goddard Space Flight Center On May, 2025 Original approval signatures are on file with the University of New Hampshire Graduate School. ACKNOWLEDGEMENTS I would like to express my heartfelt gratitude to my advisor, Dr. MD Shaad Mahmud, for his exceptional guidance, unwavering support, and insightful mentorship throughout my graduate journey at UNH. His encouragement and patience have played a pivotal role in shaping both my academic and personal growth, and for that, I am deeply thankful. Iam also profoundly grateful to my respected committee members, Dr. Diliang Chen, Dr. Donpeng Xu, Dr. Hyunju K. Connor, and Dr. Se Young Yoon, for their valuable feedback and guidance, which have significantly enriched this research. The knowledge that I have gained under their guidance will undoubtedly benefit me in both my professional and personal life. I extend my sincere appreciation to my colleagues in the Machine-learning Algorithms for GICs in Alaska and New Hampshire (MAGICIAN) team. Their collaborative spirit and intellectual contributions have made this research journey both stimulating and rewarding. I am also deeply thankful to my lab mates, Dr. Sajila Wickramaratne, MD Faishal Yousuf, Blaise O’Mara, and Sabby Clemmons, for their camaraderie, thoughtful discussions, and firm support throughout this process. I gratefully acknowledge the financial support provided by NSF EPSCOR, which was instrumental in making this research possible. This funding laid the foundation for the progress and outcomes achiev

FAQs

What problem does the dissertation address?

It targets real-time prediction of geomagnetic anomalies and geomagnetically induced currents (GICs) using robust edge machine-learning approaches.

How does the research position machine learning relative to physics-based simulations?

It presents physics-based simulation approaches and compares them with data-driven machine learning (ML)-driven methods for geomagnetic event forecasting.

What monitoring and data-processing elements are emphasized?

The work discusses geomagnetic monitoring through magnetometers and includes baseline correction as part of handling geomagnetic measurements.

Aug 4, 2026

Upload by Eden