A learning-based scheme to optimise a cognitive handoff

Simple item page
dc.contributor.advisorVentura, Necoen_ZA
dc.contributor.authorGombiro, Kurai Lukeen_ZA
dc.date.accessioned2016-07-25T11:24:31Z
dc.date.available2016-07-25T11:24:31Z
dc.date.issued2016en_ZA
dc.description.abstractThe evolution of communication standards promotes the development and use of several spectrum-sharing strategies. From the noted results, machine-learning techniques have paved a direction for radio protocols to achieve better levels of performance. With their definition, efficient learning practices and the use of effective spectrum sharing methods necessitate the development of better channel selection schemes. In this work, a radios' learning capability enables the manipulation of a spectrum-sharing concept. This involves the radio obeying certain rules in a spectrum sharing facility, which defines a decentralised form of coexistence (sharing) between the radios occupying that specific radio space. Amongst other benefits, the sharing promotes the node's independence in the radio space, between the cohabitating radios for the essence of efficient spectrum sharing. The learning dimension is realised by the use of a Stochastic Estimator Learning Automata (SELA) algorithm. It allows a radio node to roam independently, while achieving the goal of learning to control spectrum use over time. This is by selecting an effective action that defines the radio's channel choice, leading to the long-term benefit of learning the radio usage patterns. A key condition for spectrum sharing requires that a 'borrowed' channel be handed-over to the owner, in any network for the sake of fair sharing practices. The sharing practices promote the evolution of spectrum use by making use of a device called a Cognitive Radio (CR). The CR, as a device that is set to redefine the sharing landscape, creates a paradigm that will revolutionise the concept of machine learning in the communications world. For the CR to have a good level of functionality, the learning rate and evolution should be dynamic. This is because, the results from its interactions with other users enhances its capability of coexistence and further promotes the progression of the spectrum-sharing concept.en_ZA
dc.identifier.apacitationGombiro, K. L. (2016). <i>A learning-based scheme to optimise a cognitive handoff</i>. (Thesis). University of Cape Town ,Faculty of Engineering & the Built Environment ,Department of Electrical Engineering. Retrieved from http://hdl.handle.net/11427/20678en_ZA
dc.identifier.chicagocitationGombiro, Kurai Luke. <i>"A learning-based scheme to optimise a cognitive handoff."</i> Thesis., University of Cape Town ,Faculty of Engineering & the Built Environment ,Department of Electrical Engineering, 2016. http://hdl.handle.net/11427/20678en_ZA
dc.identifier.citationGombiro, K. 2016. A learning-based scheme to optimise a cognitive handoff. University of Cape Town.en_ZA
dc.identifier.ris TY - Thesis / Dissertation AU - Gombiro, Kurai Luke AB - The evolution of communication standards promotes the development and use of several spectrum-sharing strategies. From the noted results, machine-learning techniques have paved a direction for radio protocols to achieve better levels of performance. With their definition, efficient learning practices and the use of effective spectrum sharing methods necessitate the development of better channel selection schemes. In this work, a radios' learning capability enables the manipulation of a spectrum-sharing concept. This involves the radio obeying certain rules in a spectrum sharing facility, which defines a decentralised form of coexistence (sharing) between the radios occupying that specific radio space. Amongst other benefits, the sharing promotes the node's independence in the radio space, between the cohabitating radios for the essence of efficient spectrum sharing. The learning dimension is realised by the use of a Stochastic Estimator Learning Automata (SELA) algorithm. It allows a radio node to roam independently, while achieving the goal of learning to control spectrum use over time. This is by selecting an effective action that defines the radio's channel choice, leading to the long-term benefit of learning the radio usage patterns. A key condition for spectrum sharing requires that a 'borrowed' channel be handed-over to the owner, in any network for the sake of fair sharing practices. The sharing practices promote the evolution of spectrum use by making use of a device called a Cognitive Radio (CR). The CR, as a device that is set to redefine the sharing landscape, creates a paradigm that will revolutionise the concept of machine learning in the communications world. For the CR to have a good level of functionality, the learning rate and evolution should be dynamic. This is because, the results from its interactions with other users enhances its capability of coexistence and further promotes the progression of the spectrum-sharing concept. DA - 2016 DB - OpenUCT DP - University of Cape Town LK - https://open.uct.ac.za PB - University of Cape Town PY - 2016 T1 - A learning-based scheme to optimise a cognitive handoff TI - A learning-based scheme to optimise a cognitive handoff UR - http://hdl.handle.net/11427/20678 ER - en_ZA
dc.identifier.urihttp://hdl.handle.net/11427/20678
dc.identifier.vancouvercitationGombiro KL. A learning-based scheme to optimise a cognitive handoff. [Thesis]. University of Cape Town ,Faculty of Engineering & the Built Environment ,Department of Electrical Engineering, 2016 [cited yyyy month dd]. Available from: http://hdl.handle.net/11427/20678en_ZA
dc.language.isoengen_ZA
dc.publisher.departmentDepartment of Electrical Engineeringen_ZA
dc.publisher.facultyFaculty of Engineering and the Built Environment
dc.publisher.institutionUniversity of Cape Town
dc.subject.otherElectrical Engineeringen_ZA
dc.titleA learning-based scheme to optimise a cognitive handoffen_ZA
dc.typeMaster Thesis
dc.type.qualificationlevelMasters
dc.type.qualificationnameMSc (Eng)en_ZA
uct.type.filetypeText
uct.type.filetypeImage
uct.type.publicationResearchen_ZA
uct.type.resourceThesisen_ZA
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
thesis_ebe_2016_gombiro_kurai_luke.pdf
Size:
2.02 MB
Format:
Adobe Portable Document Format
Description:
Download
Collections
Masters