University of Thessaly (UTH)
Spectrum and storage capacity management using network economics and optimization methods
Abstract
dc:descriptionThe last few years we are witnessing an unprecedented growth in user demand for high speed communication and ubiquitous network access. Moreover novel communication paradigms and applications create an ever increasing volume of data traffic that must be transported in a timely and cost efficient fashion. This impressive transformation of communication networks poses new challenges for their design and management that cannot be addressed with traditional methods of resource over provisioning and technology upgrading. At the same time nowadays it is clear that critical and valuable network resources such as spectrum and in network storage are either underutilized or not adequately exploited. These observations indicate that there is still much space for improvement and call for innovative approaches in network design and network resource management. In this thesis we propose market based mechanisms and optimization methods in order to increase utilization of critical network resources and improve the performance of networks in terms of data transfer capability. These mechanisms are of paramount importance for contemporary networks which are often plagued by the lack of coordination and the egotistic behavior of their constituent nodes. First we focus on spectrum management a resource that is scarce and at the same time remains underutilized in large extent. A prominent proposed solution for this problem is the reform of spectrum allocation policy and the deployment of dynamic spectrum (DS) markets. In these markets spectrum will be freely traded as a commodity and very often hierarchical structures among the interacting entities will emerge. We study these multilayer markets and propose a mechanism that increases the allocative efficiency of spectrum. We consider the representative scenario that arises when a governmental agency sells spectrum channels to Primary Operators (POs) who subsequently resell them to Secondary Operators (SOs) through auctions in monopolistic markets. We show that this hierarchical scheme does not ensure a socially efficient spectrum allocation which is aimed by the agency due to lack of coordination among the entities in different layers and the selfish revenue maximizing strategy of POs. In order to reconcile these opposing objectives we propose a pricing based incentive mechanism which aligns the actions of the POs with the objective of the agency and thus leads to system performance improvement in terms of social welfare. The suggested mechanism constitutes a method for regulation which is proved to be of crucial importance for the emerging hierarchical spectrum markets. Next we relax the assumption of monopolistic markets and analyze the competition of wireless services providers over a common pool of users. We show that lack of information about the actual network capacity and egotistic strategies of operators may induce revenue reduction for the latter or efficiency loss for the market. We assume that users have a reservation utility or equivalently an alternative option to satisfy their communication needs. The operators must satisfy these minimum requirements in order to attract clients. We model the interaction of users as an evolutionary game and the competition among the operators as a non cooperative pricing game. We prove that the equilibriums of both games depend on the reservation utility and the amount of spectrum the operators have at their disposal. Accordingly we consider the scenario where a regulating agency is able to intervene in the market by tuning these parameters. Different regulators may have different objectives and criteria. We adopt a mechanism design perspective analyze the various possible regulation methods and discuss their requirements implications and impact on the welfare of the market and the revenue of the operators. Apart from the hierarchical spectrum allocation through monopolistic or oligopolistic markets another crucial aspect of these dynamic spectrum management schemes is that they aim to facilitate spectrum exchange among different entities in the same layer. For example each secondary operator may lease or even exchange its channels to other operators. Similarly users will be able to exchange bandwidth in order to satisfy their dynamic needs. In this setting each entity is both a resource provider and a resource consumer. Moreover these two roles are intertwined since they both presume the consumption of the entity’s scarce spectrum resource. We model this problem and propose a dynamic pricing scheme that clears the market and ensures the maximization of social Welfare. Through the adoption of proper pricing and allocation rules selfish behavior is deterred and the actual needs of the market entities are revealed and satisfied. Market based network management methods provide incentives to network entities for increasing network resource utilization. Equally important for improving network performance is the exploitation of resources which until now have not been incorporated in the network design. A prominent example is in network storage a resource that nowadays is very cheap compared to bandwidth and available at large scale. We advocate that storage can be used to improve the data transfer capability of dynamic networks i.e. networks with time varying link capacities and we identify the conditions under which this improvement is realizable. The basic idea is to temporarily store data at nodes when network conditions (e.g. link capacities) are not favorable for data transmission. We use the technique of time expanded graphs in order to map the dynamic networks to equivalent static networks. First we introduce a method that increases the mm cut of the network by adding storage to certain nodes. Next we propose the conjunction of storage control with routing. We define the max flow problem in the time expanded graph and solve it by using a distributed algorithm. The solution constitutes the optimal joint storage control and routing (JSR) policy. Finally we discuss the importance of available information about current and future network state on performance of the proposed algorithms.
Degree
thesis:*- Grantor dc:publisher
- University of Thessaly (UTH)
- Year dc:date
- 2012
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Iosifidis, George
- Ιωσηφίδης, Γεώργιος
Subjects
dc:subject × 16- Διαχείριση αποθηκευτικών χώρων
- Δημοπρασίες
- Διαχείριση φάσματος
- Θεωρία παιγνίων
- Κοστολόγηση
- Μηχανισμοί παροχής κινήτρων
- Storage capacity managment
- Auctions
- Spectrum management
- Game theory
- Pricing
- Incentive mechanisms
- Φυσικές Επιστήμες
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Natural Sciences
- Computer and Information Sciences
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/26952