Date of Award

3-2025

Document Type

Thesis

Degree Name

Master of Science in Electrical Engineering

Department

Department of Electrical and Computer Engineering

First Advisor

Nicholas J. Yielding, PhD

Abstract

This research investigates jamming evasion using DRL to provide an autonomous solution that repositions a geostationary satellite experiencing directed, terrestrial based jamming. Second, this thesis applies DRL to an area of research for LEO satellite constellations, user grouping, using a portion of an Air Force Research Lab reinforcement learning framework. As LEO satellites orbit the earth, they must constantly re-evaluate not only which users they are able to connect to, but on which of its multiple beams. This model succeeds in finding a balance between maximizing signal strength and minimizing overhead from switching user assignments, all while requiring fewer costly channel state information calculations when compared to a baseline method of user grouping.

AFIT Designator

AFIT-ENG-MS-25-M-032

Comments

An embargo was observed for posting this thesis on AFIT Scholar.

Approved for public release, distribution unlimited. PA case number 88ABW-2025-0274.

AFIT Scholar Admin note: In accordance with clearance stipulations in the PA case cited above, a required substitute disclaimer was placed on the document at pre-page ii (page 2 of the PDF):
"The views expressed are those of the authors and do not reflect the official guidance or position of the United States Government, the Department of Defense, the United States Air Force or the United States Space Force."

Download

Share

COinS