{"id":{"repo_id":"soton","oai_identifier":"oai:eprints.soton.ac.uk:512167"},"canonical_url":"https://search.ndltd.org/etd/soton/oai:eprints.soton.ac.uk:512167","repository":{"repo_id":"soton","name":"University of Southampton","base_url":"https://eprints.soton.ac.uk/cgi/oai2"},"display":{"title":"Optimising cooperative non-orthogonal multiple access and hierarchical modulation","abstract":"This thesis investigates the optimisation of Cooperative Non-Orthogonal Multiple Access (C-NOMA) and Hierarchical Modulation (HM) for next-generation wireless networks. Both techniques exploit superposition coding to enhance spectral efficiency and user fairness, yet each presents distinct challenges. The thesis presented here addresses some of the resource allocation challenges experienced through the application of game theory and Reinforcement Learning (RL) for various cooperative networks.<br/><br/>Game-theoretic approaches were employed to address user pairing and power allocation in multi-carrier C-NOMA networks. By formulating the resource allocation problem as a distributed game, two algorithms were proposed: Conventional Distributed Algorithm (CDA) and Pragmatic Distributed Algorithm (PDA). These methods enabled users to make independent yet coordinated decisions, ensuring that fairness and Quality of Service (QoS) constraints were respected while improving overall spectral efficiency. Crucially, this distributed design reduced reliance on centralised optimisation and provided a reduced computation complexity compared to maximum likelihood searches.<br/><br/>Deep Reinforcement Learning (DRL) was then applied to adaptive cooperative HM under imperfect channel state information. A dual-agent proximal policy optimisation framework was introduced, where one agent controlled modulation and coding schemes while the other allocated transmission power and constellation parameters. This division of tasks allowed the system to dynamically adapt to channel estimation error and various deployments of relays whilst maintaining an acceptable Bit Error Rate (BER). Results showed that reinforcement learning provides a robust and efficient solution to managing the complexity of HM-based cooperative networks.<br/><br/>Finally, HM and C-NOMA were integrated to explore hybrid cooperative schemes. Exploitation of both systems in unison, even without optimising the challenges aforementioned, showed to provide greater throughput by employing both systems approaches to the cooperative relaying of messages.","abstract_html":"This thesis investigates the optimisation of Cooperative Non-Orthogonal Multiple Access (C-NOMA) and Hierarchical Modulation (HM) for next-generation wireless networks. Both techniques exploit superposition coding to enhance spectral efficiency and user fairness, yet each presents distinct challenges. The thesis presented here addresses some of the resource allocation challenges experienced through the application of game theory and Reinforcement Learning (RL) for various cooperative networks.&lt;br/&gt;&lt;br/&gt;Game-theoretic approaches were employed to address user pairing and power allocation in multi-carrier C-NOMA networks. By formulating the resource allocation problem as a distributed game, two algorithms were proposed: Conventional Distributed Algorithm (CDA) and Pragmatic Distributed Algorithm (PDA). These methods enabled users to make independent yet coordinated decisions, ensuring that fairness and Quality of Service (QoS) constraints were respected while improving overall spectral efficiency. Crucially, this distributed design reduced reliance on centralised optimisation and provided a reduced computation complexity compared to maximum likelihood searches.&lt;br/&gt;&lt;br/&gt;Deep Reinforcement Learning (DRL) was then applied to adaptive cooperative HM under imperfect channel state information. A dual-agent proximal policy optimisation framework was introduced, where one agent controlled modulation and coding schemes while the other allocated transmission power and constellation parameters. This division of tasks allowed the system to dynamically adapt to channel estimation error and various deployments of relays whilst maintaining an acceptable Bit Error Rate (BER). Results showed that reinforcement learning provides a robust and efficient solution to managing the complexity of HM-based cooperative networks.&lt;br/&gt;&lt;br/&gt;Finally, HM and C-NOMA were integrated to explore hybrid cooperative schemes. Exploitation of both systems in unison, even without optimising the challenges aforementioned, showed to provide greater throughput by employing both systems approaches to the cooperative relaying of messages.","abstract_has_math":false,"creators":["Horler, Harry"],"institution":"University of Southampton","degree_name":"Ph.D.","degree_level":"doctoral","degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":["Ng, Michael","Rastegari, Bahar"],"committee_chairs":[],"committee_members":[],"year":2026,"date_issued":"2026","date_published":"2026","updated_at":"2026-11-19T09:55:51Z","subjects":[],"languages":["English","eng"],"rights":[],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[{"key":"person:identifier.orcid","label":"ORCID","values":["000900052714552X","0000000209307194","000000020985573X"],"render_values":[{"text":"0009-0005-2714-552X","href":"https://orcid.org/0009-0005-2714-552X","code":true},{"text":"0000-0002-0930-7194","href":"https://orcid.org/0000-0002-0930-7194","code":true},{"text":"0000-0002-0985-573X","href":"https://orcid.org/0000-0002-0985-573X","code":true}]}]},"links":{"outbound_url":"https://doi.org/10.5258/SOTON/01/T207","outbound_label":"DOI","outbound_source":"dc:identifier.doi","record_url":"https://doi.org/10.5258/SOTON/01/T207","record_label":"Repository record","record_kind":"doi","record_source":"dc:identifier.doi","download_url":"https://eprints.soton.ac.uk/id/eprint/512167/1/PhD_Thesis.pdf","download_label":"Download PDF","download_source":"dc:identifier.uri","access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://eprints.soton.ac.uk/cgi/oai2?verb=GetRecord&metadataPrefix=uketd_dc&identifier=oai%3Aeprints.soton.ac.uk%3A512167","prefix":"uketd_dc"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Ng, Michael","Rastegari, Bahar"]},{"key":"dc:creator","label":"Author","values":["Horler, Harry"]},{"key":"person:identifier.orcid","label":"ORCID","values":["000900052714552X","0000000209307194","000000020985573X"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2026"]},{"key":"dc:date.issued","label":"Date","values":["2026"]},{"key":"dc:publisher.commercial","label":"Dc Publisher Commercial","values":["University of Southampton"]},{"key":"dc:publisher.department","label":"Dc Publisher Department","values":["School of Electronics and Computer Science"]},{"key":"dc:publisher.institution","label":"Dc Publisher Institution","values":["University of Southampton"]},{"key":"dc:relation.isreferencedby","label":"Dc Relation Isreferencedby","values":["https://eprints.soton.ac.uk/id/eprint/512167/"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"dc:type.qualificationlevel","label":"Dc Type Qualificationlevel","values":["doctoral"]},{"key":"dc:type.qualificationname","label":"Dc Type Qualificationname","values":["Ph.D."]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["English"]},{"key":"dc:language.dcterms_iso639_2","label":"Language (ISO 639-2)","values":["eng"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.doi","label":"DOI","values":["10.5258/SOTON/01/T207"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://eprints.soton.ac.uk/id/eprint/512167/1/PhD_Thesis.pdf","https://eprints.soton.ac.uk/id/eprint/512167/2/Final-thesis-submission-Examination-Mr-Harry-Horler.pdf"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["This thesis investigates the optimisation of Cooperative Non-Orthogonal Multiple Access (C-NOMA) and Hierarchical Modulation (HM) for next-generation wireless networks. Both techniques exploit superposition coding to enhance spectral efficiency and user fairness, yet each presents distinct challenges. The thesis presented here addresses some of the resource allocation challenges experienced through the application of game theory and Reinforcement Learning (RL) for various cooperative networks.<br/><br/>Game-theoretic approaches were employed to address user pairing and power allocation in multi-carrier C-NOMA networks. By formulating the resource allocation problem as a distributed game, two algorithms were proposed: Conventional Distributed Algorithm (CDA) and Pragmatic Distributed Algorithm (PDA). These methods enabled users to make independent yet coordinated decisions, ensuring that fairness and Quality of Service (QoS) constraints were respected while improving overall spectral efficiency. Crucially, this distributed design reduced reliance on centralised optimisation and provided a reduced computation complexity compared to maximum likelihood searches.<br/><br/>Deep Reinforcement Learning (DRL) was then applied to adaptive cooperative HM under imperfect channel state information. A dual-agent proximal policy optimisation framework was introduced, where one agent controlled modulation and coding schemes while the other allocated transmission power and constellation parameters. This division of tasks allowed the system to dynamically adapt to channel estimation error and various deployments of relays whilst maintaining an acceptable Bit Error Rate (BER). Results showed that reinforcement learning provides a robust and efficient solution to managing the complexity of HM-based cooperative networks.<br/><br/>Finally, HM and C-NOMA were integrated to explore hybrid cooperative schemes. Exploitation of both systems in unison, even without optimising the challenges aforementioned, showed to provide greater throughput by employing both systems approaches to the cooperative relaying of messages."]},{"key":"dc:format","label":"Dc Format","values":["text"]},{"key":"dcterms:accessrights","label":"Dcterms Accessrights","values":["public","staffonly"]},{"key":"dc:title","label":"Title","values":["Optimising cooperative non-orthogonal multiple access and hierarchical modulation"]}]}],"canonical_facts":{"dc:contributor.advisor":["Ng, Michael","Rastegari, Bahar"],"dc:creator":["Horler, Harry"],"dc:date":["2026"],"dc:date.issued":["2026"],"dc:description.abstract":["This thesis investigates the optimisation of Cooperative Non-Orthogonal Multiple Access (C-NOMA) and Hierarchical Modulation (HM) for next-generation wireless networks. Both techniques exploit superposition coding to enhance spectral efficiency and user fairness, yet each presents distinct challenges. The thesis presented here addresses some of the resource allocation challenges experienced through the application of game theory and Reinforcement Learning (RL) for various cooperative networks.<br/><br/>Game-theoretic approaches were employed to address user pairing and power allocation in multi-carrier C-NOMA networks. By formulating the resource allocation problem as a distributed game, two algorithms were proposed: Conventional Distributed Algorithm (CDA) and Pragmatic Distributed Algorithm (PDA). These methods enabled users to make independent yet coordinated decisions, ensuring that fairness and Quality of Service (QoS) constraints were respected while improving overall spectral efficiency. Crucially, this distributed design reduced reliance on centralised optimisation and provided a reduced computation complexity compared to maximum likelihood searches.<br/><br/>Deep Reinforcement Learning (DRL) was then applied to adaptive cooperative HM under imperfect channel state information. A dual-agent proximal policy optimisation framework was introduced, where one agent controlled modulation and coding schemes while the other allocated transmission power and constellation parameters. This division of tasks allowed the system to dynamically adapt to channel estimation error and various deployments of relays whilst maintaining an acceptable Bit Error Rate (BER). Results showed that reinforcement learning provides a robust and efficient solution to managing the complexity of HM-based cooperative networks.<br/><br/>Finally, HM and C-NOMA were integrated to explore hybrid cooperative schemes. Exploitation of both systems in unison, even without optimising the challenges aforementioned, showed to provide greater throughput by employing both systems approaches to the cooperative relaying of messages."],"dc:format":["text"],"dc:identifier.doi":["10.5258/SOTON/01/T207"],"dc:identifier.uri":["https://eprints.soton.ac.uk/id/eprint/512167/1/PhD_Thesis.pdf","https://eprints.soton.ac.uk/id/eprint/512167/2/Final-thesis-submission-Examination-Mr-Harry-Horler.pdf"],"dc:language":["English"],"dc:language.dcterms_iso639_2":["eng"],"dc:publisher.commercial":["University of Southampton"],"dc:publisher.department":["School of Electronics and Computer Science"],"dc:publisher.institution":["University of Southampton"],"dc:relation.isreferencedby":["https://eprints.soton.ac.uk/id/eprint/512167/"],"dcterms:accessrights":["public","staffonly"],"dc:title":["Optimising cooperative non-orthogonal multiple access and hierarchical modulation"],"dc:type":["Thesis"],"dc:type.qualificationlevel":["doctoral"],"dc:type.qualificationname":["Ph.D."],"person:identifier.orcid":["000900052714552X","0000000209307194","000000020985573X"]},"updated_at":"2026-11-19T09:55:51Z"}