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

Show simple item record

dc.contributor.advisor Maharaj, Sunil
dc.contributor.postgraduate Sande, Malcolm Makomborero
dc.date.accessioned 2023-02-27T10:58:42Z
dc.date.available 2023-02-27T10:58:42Z
dc.date.created 2023-05-12
dc.date.issued 2023
dc.description Thesis (PhD (Electronic Engineering))--University of Pretoria, 2023.. en_US
dc.description.abstract The 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.availability Unrestricted en_US
dc.description.degree PhD (Electronic Engineering) en_US
dc.description.department Electrical, Electronic and Computer Engineering en_US
dc.description.sponsorship Sentech Chair in Broadband Wireless Multimedia Communications. en_US
dc.identifier.citation * en_US
dc.identifier.doi https://doi.org/10.25403/UPresearchdata.22182295 en_US
dc.identifier.other A2023
dc.identifier.uri https://repository.up.ac.za/handle/2263/89854
dc.language.iso en en_US
dc.publisher University 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.subject UCTD en_US
dc.subject Wireless Communications en_US
dc.subject Machine learning en_US
dc.subject Congestion control
dc.subject Deep reinforcement learning
dc.subject Integrated access and backhaul
dc.title Resource Management and Backhaul Routing in Millimeter-Wave IAB Networks Using Deep Reinforcement Learning en_US
dc.type Thesis en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record