Anchora, Luca (2012) Spectrum sharing in mobile cellular networks: an alternative approach for efficient resource utilization. Advisor: Badia, Dr. Leonardo. pp. 125. [IMT PhD Thesis]
|
Text
Anchora_phdthesis.pdf - Published Version Available under License Creative Commons Attribution No Derivatives. Download (10MB) | Preview |
Abstract
Mobile cellular communications have been becoming the leading technology in data transmissions: trillions of devices will serve billions of people in few years, not only for simple phone calls but also for application data transfer. The rapid diffusion of this technology goes with an increasing use of resources, in particular of bandwidth, whose scarcity and expensiveness make efficient management necessary. To this aim, a dynamic allocation of the spectrum is preferred to an inefficient static allocation. In this way, the waste of resources is reduced by assigning the unused frequencies to nodes that need them. Many flexible ways to use the resource can be thought of, provided that the quality of service requirements of each user are respected. In this thesis the problem of efficient spectrum usage is considered in a two-fold manner, namely considering the intra-cell and the inter-cell context. In the former, the need for an efficient resource utilization in the downlink is traded-off with fairness among the user flows, so a possible system model is discussed and some allocation algorithms are proposed. Two algorithms for the dynamic setting of a structural parameter are proposed and validated thus showing the efficiency of the operating point into which the radio resource manager is lead. Regarding the inter-cell context, a particular type of scenario is considered, a multi-operator cellular network. In this case, an alternative to the classical static frequency allocation is proposed: spectrum sharing. By allowing operators to share part of the spectrum that they receive by the regulation body, a gain in terms of cell throughput can be achieved thanks to a better utilization of the shared resources. In this case, a crucial issue is the access mechanism to the common spectrum. Two algorithms that define the upper bound and the lower bound on the system performance are given. Another possible mechanism that takes into consideration priorities on the access to the common resources is analyzed as well. The main result is that there is an asymptotic gain for the operators in sharing their resources, which is a fundamental point for the diffusion of this proposal. This opens the door to the implementation of effective algorithms of spectrum sharing in scenarios with realistic constraints. Moreover, the main factors that impact on the value of such a gain have been identified and analyzed. Most of the mathematical models considered all over this thesis are based on the Game Theory. Multi-agent systems perfectly fit such a framework, through which equilibrium points and efficiency of solutions can be evaluated together with possible cooperative strategies. However, in many situations the intrinsic complexity of the sytem or the consideration of realistic scenarios may make the analytical treatment tough. Therefore, the validation through simulations is important as well. A modular framework, obtained through the extension of a well-known simulation platform, has been developed for the support to the scenarios of interest and has been used for the validation of all the proposed algorithms. The details of its software architecture are given as well.
| Item Type: | IMT PhD Thesis |
|---|---|
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science |
| PhD Course: | Computer Science and Engineering |
| Identification Number: | 10.6092/imtlucca/e-theses/33 |
| NBN Number: | urn:nbn:it:imtlucca-27069 |
| Date Deposited: | 10 Jul 2012 14:26 |
| URI: | http://e-theses.imtlucca.it/id/eprint/33 |
Actions (login required, only for staff repository)
 |
View Item |
