Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.777589
Title: Spectrum aggregation for dynamic spectrum access based cognitive networks
Author: Rostami, Soheil
ISNI:       0000 0004 7963 366X
Awarding Body: University of Greenwich
Current Institution: University of Greenwich
Date of Award: 2016
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
Telecommunication standards bodies and vendors all over the world are seeking ways of continuing the evolution of mobile communication technology on to its next generation. Providing ubiquitous broadband access to all mobile users distributed in a given area with a fragmented spectrum remains a challenging task. This thesis contributes to the global efforts by extending bandwidth by means of Spectrum Aggregation (SA) in Next Generation Mobile Networks (NGMN). First of all, the motivation behind SA for NGMN is investigated. SA is introduced to create large virtual carrier bandwidths for data hungry users by aggregating fragmented segments of spectrum. By using SA, multiple carriers with different bandwidths, dispersed within intra or inter-bands, can be simultaneously utilised to provide higher data rates, better coverage and simplified multi-band traffic management, resulting in an enhanced user quality of experience. However, SA functionality introduces new challenges for transceiver architecture and Radio Resource Management (RRM) functionality of the network. Secondly, spectrum assignment for cognitive radio networks with SA is investigated. For this purpose, an aggregation-based spectrum assignment is formulated as an integer optimisation problem. The problem is solved using a genetic algorithm to exploit the discontinuity of the available spectrum and to achieve higher capacity gains. The formulation not only considers interference to primary users but also takes into account co-channel interference among secondary users and maximum aggregation span. Simulation results clearly show that the proposed algorithm outperforms state-of-the-art solutions in terms of spectrum utilisation and higher convergence rate. Thirdly, RRM in Long-Term Evolution (LTE) networks with SA is considered. It is shown that the optimal solution of RRM in different scenarios can be achieved by solving the relaxed optimisation problem. The optimum RRM algorithm with reduced complexity is proposed and compared with the current solutions in literature. The proposed algorithm optimally assigns component carriers, resource blocks, modulation and coding scheme values based on users' channel state information and SA capabilities. Simulation results clearly show that the proposed algorithm outperforms current solutions in terms of cell throughput and user throughput fairness. Last but not least, resource allocation with SA for orthogonal frequency-division multiplexing based networks is studied. A joint resource allocation problem with SA functionality is formulated and sub-optimal solutions with low computational complexity are proposed. Simulation results verify that the proposed algorithms achieve near optimum throughput and exploit the multi-carrier diversity of wireless systems.
Supervisor: Arshad, Kamran ; Rapajic, Predrag Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.777589  DOI: Not available
Keywords: TK Electrical engineering. Electronics Nuclear engineering
Share: