Improving Radar Sensing Capabilities and Data Quality Through Machine Learning

Amaireh, Anas

Improving Radar Sensing Capabilities and Data Quality Through Machine Learning

dc.contributor.advisor	Zhang, Yan
dc.contributor.author	Amaireh, Anas
dc.contributor.committeeMember	Homeyer, Cameron
dc.contributor.committeeMember	Schvartzman, David
dc.contributor.committeeMember	Cheng, Samuel
dc.contributor.committeeMember	Ebert, David
dc.date.accessioned	2024-09-24T16:19:24Z
dc.date.available	2024-09-24T16:19:24Z
dc.date.issued	2024-12-13
dc.date.manuscript	2024-09-05
dc.description.abstract	This dissertation integrates advanced machine learning (ML) techniques with radar technology to address significant challenges in atmospheric sciences, cloud profiling, and aviation safety. It aims to enhance the accuracy and reliability of radar-based measurements, improve the prediction of atmospheric relative humidity and Cloud Liquid Water Content (CLWC), and mitigate the impact of 5G interference on radar altimeters. These improvements are essential for advancing public safety, weather forecasting, and aviation technology. Chapter 2 provides a comprehensive overview of ML, detailing its history, evolution, and significance in scientific research. It introduces supervised, unsupervised, and reinforcement learning, and discusses various ML models, such as regression and classification, establishing a foundation for integrating ML with radar technology. Chapter 3 introduces a method for estimating atmospheric relative humidity using wind profiler radar and a cascaded ML algorithm. Unlike existing methods, this approach uses only moment data to generate an intermediate pressure profile, serving as training data for humidity estimations without requiring temperature as an input feature. The study evaluates various ML algorithms using radiosonde data from the Hong Kong Observatory, demonstrating the effectiveness of this simplified, feature-efficient model. Chapter 4 uses ML techniques to enhance Cloud Liquid Water Content (CLWC) profiling. The study cross-validates ERA5 data with high-precision radiosonde observations from Hong Kong. It employs ML to interpolate radiosonde data to improve coverage and resolution, and uses a metaheuristic algorithm to cleanse data. This enhances the correlation between input features and CLWC. The results show significant improvements in the accuracy and reliability of CLWC profile prediction. Chapter 5 addresses the critical issue of 5G interference with radar altimeter signals, which is crucial for aviation safety. A new ML framework is developed to classify signals into pure or interfered categories and predict altitudes when interference is detected. The study employs real 5G signals from a base station in Norman, Oklahoma, and emulated radar signals to train and test the framework. This approach ensures the accuracy of radar altimeters despite the level of 5G interference. This dissertation demonstrates the transformative potential of integrating ML techniques with radar technology. The proposed solutions enhance radar sensing capabilities and data quality, significantly improving public safety, weather forecasting, and aviation.	en_US
dc.identifier.uri	https://hdl.handle.net/11244/340632
dc.language	en_US	en_US
dc.rights	Attribution-NonCommercial-NoDerivatives 4.0 International	*
dc.rights.uri	http://creativecommons.org/licenses/by-nc-nd/4.0/	*
dc.subject	Machine learning	en_US
dc.subject	Radar	en_US
dc.subject	Telecommunications	en_US
dc.subject	Aviation Safety	en_US
dc.thesis.degree	Ph.D.	en_US
dc.title	Improving Radar Sensing Capabilities and Data Quality Through Machine Learning	en_US
ou.group	Gallogly College of Engineering::School of Electrical and Computer Engineering	en_US
shareok.orcid	https://orcid.org/0000-0002-2721-4872	en_US