Deep learning-based radio frequency interference detection and mitigation for microwave radiometers with 2-D spectral features

Author

Ahmed Manavi Alam, Mississippi State UniversityFollow

Advisor

Gurbuz, Ali Cafer

Committee Member

Kurum, Mehmet

Committee Member

Marojevic, Vuk

Date of Degree

8-13-2024

Original embargo terms

Immediate Worldwide Access

Document Type

Graduate Thesis - Open Access

Major

Electrical & Computer Engineering

Degree Name

Master of Science (M.S.)

College

James Worth Bagley College of Engineering

Department

Department of Electrical and Computer Engineering

Abstract

Radio frequency interference (RFI) poses significant challenges for passive microwave radiometry used in climate studies and Earth science. Despite operating in protected frequency bands, microwave radiometers often encounter RFI from sources like air surveillance radars, 5G communications, and unmanned aerial vehicles. Traditional RFI detection methods rely on handcrafted algorithms designed for specific RFI types. This study proposes a deep learning (DL) approach, leveraging convolutional neural networks to detect various RFI types on a global scale. By learning directly from radiometer data, this data-driven method enhances detection accuracy and generalization. The DL framework processes raw moment data and Stokes parameters, dynamically labeled using quality flags, offering a robust and efficient solution for RFI detection. This approach demonstrates the potential for improved RFI mitigation in passive remote sensing applications.

Recommended Citation

Alam, Ahmed Manavi, "Deep learning-based radio frequency interference detection and mitigation for microwave radiometers with 2-D spectral features" (2024). Theses and Dissertations. 6215.
https://scholarsjunction.msstate.edu/td/6215