Resource Management and Backhaul Routing in Millimeter-Wave IAB Networks Using Deep Reinforcement Learning

dc.contributor.advisorMaharaj, Sunil
dc.contributor.emailmalcolm.sande@gmail.comen_US
dc.contributor.postgraduateSande, Malcolm Makomborero
dc.date.accessioned2023-02-27T10:58:42Z
dc.date.available2023-02-27T10:58:42Z
dc.date.created2023-05-12
dc.date.issued2023
dc.descriptionThesis (PhD (Electronic Engineering))--University of Pretoria, 2023..en_US
dc.description.abstractThe increased densification of wireless networks has led to the development of integrated access and backhaul (IAB) networks. In this thesis, deep reinforcement learning was applied to solve resource management and backhaul routing problems in millimeter-wave IAB networks. In the research work, a resource management solution that aims to avoid congestion for access users in an IAB network was proposed and implemented. The proposed solution applies deep reinforcement learning to learn an optimized policy that aims to achieve effective resource allocation whilst minimizing congestion and satisfying the user requirements. In addition, a deep reinforcement learning-based backhaul adaptation strategy that leverages a recursive discrete choice model was implemented in simulation. Simulation results where the proposed algorithms were compared with two baseline methods showed that the proposed scheme provides better throughput and delay performance.en_US
dc.description.availabilityUnrestricteden_US
dc.description.degreePhD (Electronic Engineering)en_US
dc.description.departmentElectrical, Electronic and Computer Engineeringen_US
dc.description.sponsorshipSentech Chair in Broadband Wireless Multimedia Communications.en_US
dc.identifier.citation*en_US
dc.identifier.doihttps://doi.org/10.25403/UPresearchdata.22182295en_US
dc.identifier.otherA2023
dc.identifier.urihttps://repository.up.ac.za/handle/2263/89854
dc.language.isoenen_US
dc.publisherUniversity of Pretoria
dc.rights© 2022 University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria.
dc.subjectUCTDen_US
dc.subjectWireless Communicationsen_US
dc.subjectMachine learningen_US
dc.subjectCongestion control
dc.subjectDeep reinforcement learning
dc.subjectIntegrated access and backhaul
dc.titleResource Management and Backhaul Routing in Millimeter-Wave IAB Networks Using Deep Reinforcement Learningen_US
dc.typeThesisen_US

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Sande_Resource_2023.pdf
Size:
8.22 MB
Format:
Adobe Portable Document Format
Description:

License bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
1.75 KB
Format:
Item-specific license agreed upon to submission
Description: