Order PDf Issue

Download This Issue PDF

Volume 4, No. 3, July 2008        ISSN: 1744-2397

A SPECIAL ISSUE Recent Advances In Heterogeneous Cognitive
 Wireless Networks

Guest Editors:

Prof. Anthony C. Boucouvalas, Dr. Alexandros Kaloxylos
Department of Telecommunications Science and Technology,
 University of Peloponnese, Greece

Dr. Didier Bourse
Motorola Labs Parc Les Algorithmes, Saint Aubin, France

Authors: C. Anagnostopoulos, S. Hadjiefthymiades, E. Zervas, Y. Ntarladimas,......pp. 91-98
Abstract: Bio-inspired computing models have gained significant popularity in the engineering of distributed and autonomous systems. Epidemic models assist in the direction of collaborative computing where groups of nodes can collectively share contextual information (context) based on the current mobile behavior and user interests. Such models greatly resemble the spreading of a virus in the biological perspective (epidemics). This abstraction allows us to design and implement context spreading schemes with increased efficiency. The pieces of context are represented through a contextual hierarchy induced by semantic dependencies. This is reflected in the epidemical framework through the ideas of virus transmutation, which is included in our framework and related study. Disseminated information received by a node could infect the node and improve the existing knowledge according to pre-existing semantic structure. The capability of a node to reason about context results in locally inferred information, which becomes a new potential epidemic. The application of epidemiology and context-awareness on semantically enriched information dissemination in ad-hoc mobile networks is analyzed and assessed using simulations.
Keywords: Ad-hoc network information dissemination, Autonomous Systems, Context awareness, Epidemics.

Download Article PDF

Authors: Y. Bai, L. Chen,.....................................................................................pp. 99-114
Abstract:  Recently flexible spectrum allocation between Radio Access Networks (RANs) has been considered to expand the spectrum usage enabled by Cognitive Radio (CR) technologies. With this evolution, the mobile users can freely select and adapt to his/her preferred serving RAN, and the RAN operators can dynamically allocate spectrum through spectrum broker to meet the service demand from mobile users for profit maximization. In such a networking environment, the RAN operators need to compete with each other in acquiring the spectrum from a common spectrum pool and also in attracting mobile users from a common user base. To understand the interactions and competitions between the involved parties, we formulate a sequential two-stage non-cooperative game model consisting of a spectrum allocation competition in the first stage and a service price competition in the second stage. Then we derive the Nash equilibria and their existence conditions of the subgames in each stage and the entire two-stage game. Our analysis reveals that the entire two-stage game ends up with the amount of allocated spectrum and the service price that are decided in the first stage. Next, we present and interpret the numerical results of the network capacities and the earned profits of RANs at the Nash equilibrium. Furthermore, we discuss the effects of influential factors of spectrum broker (cost of spectrum allocation), RAN operators (spectrum efficiency), and users (willing-to-pay price) on the competition results.
Keywords: Cognitive Radio, Spectrum Allocation, Game Theory.

Download Article PDF

Authors: V. Gazis, N. Alonistioti, L. Merakos,....................................................pp. 115-124
Abstract: At the dawn of the new millennium, several wireless mobile access systems, such as cellular wireless systems (e.g., GSM, GPRS, EDGE, UMTS, cdma2000, and others), broadband WLAN type systems (e.g., IEEE 802.11a/b/g), fixed wireless access systems (e.g., IEEE 802.16), broadcast systems (e.g., DAB, DVB-T) and short-range connectivity systems (e.g., Bluetooth) have become available while new ones are being developed rapidly. It is commonly understood that, in the near future, most (if not all) of these systems will be enhanced by the introduction of artificial intelligence capacities that will give rise to cognitive wireless networks. This article focuses on the knowledge management aspects of cognitive wireless networks, in particular, the need to develop suitable ontologies to efficiently support the envisaged cognitive capacity. To this end, we present a generic ontology to describe the adaptation options of the dynamic protocol stacks that will reside in future communication devices. We identify the design requirements of describing such options and elaborate on our ontology’s design rationale. Furthermore, we apply our ontology against modern protocol stacks developed in the 3GPP standardization efforts for cellular mobile networks. In addition, we develop and present an algorithm for the discovery of all feasible protocol layer and protocol stack combinations that render an integral communication subsystem realizing a specific set of services. In addition we present the logical architecture of a respective prototype implementation invested with cognitive reasoning capacities which we trial in discovering all feasible protocol stack combinations for the protocol stacks defined by the GPRS, UMTS and IMS standards. In doing so, we present performance figures for the discovery process and elaborate on our findings. Finally, we summarize the lessons learned from these trials and provide directions for future work.
Keywords: Autonomic, Cognitive, Ontology, Protocol Stack, Reconfiguration, Mobile.

Download Article PDF

Authors: L. Kovacs, A. Vidacs, B. Heder,.........................................................pp. 125-138
Abstract: The radio spectrum is a scarce, valuable and thus expensive resource. An efficient use of frequencies is primordial, but existing management systems fail to achieve that. On one hand, a substantial amount of spectrum is wasted at a given time and place with today’s rigid frequency usage policies. On the other hand, spectrum usage rights are not always given to those who have the greatest demand and would benefit from it the best, eventuating in lower overall social welfare. The possibility of on-demand spectrum allocation with Dynamic Spectrum Access (DSA) methods, together with the pressure of a liberalized spectrum market would lead to a much more efficient spectrum usage. In this paper we propose a centralized spectrum auction method in a DSA framework to distribute spectrum for fixed infrastructure wireless networks according to the needs of providers. The conditions for a feasible spectrum allocation are given, where the interference issues between different providers and different regions are handled. To achieve a feasible, efficient and optimal allocation, a one-shot, multi-bid auction method together with a second-price pricing scheme is proposed. Since tolerance in a DSA scenario is much rewarded, our pricing scheme charges providers who do not tolerate others and interfere to a larger extent than necessary. Heuristics are also given to find the optimal frequency allocations with minimal overall interference. Simulation results are presented to validate our proposal and highlight the achievable gains.
Keywords: Dynamic Spectrum Allocation, Wireless Network, Auction, Pricing, Interference-Tolerant.

Download Article PDF

Authors: A. Kaloxylos, M. Stamatelatos, A. Boucouvalas,..................................pp. 139-147
Abstract: Current technological advances have already enabled users to select their services from several providers using different access technologies. What is currently missing is the capability to move from one type of network to another seamlessly and having the terminal’s performance optimized for the serving network’s environment. The obstacle for achieving this goal is the plethora of different network architectures, protocols, and technologies that have been proposed and specified. In this paper we focus on how reconfigurability and a cross-layered terminal architecture can achieve the aforementioned target as far as mobility management is concerned. More specifically, our solution is based on the usage of IEEE 802.21 protocol to enable the communication between mobility protocols with the MAC protocols. We also adopt a reconfigurable protocol stack that enables the adaptation of the MT in every network environment. The architecture is specified using UML and detailed examples of the terminal operation are also provided.
Keywords: Mobility management, Handover, Reconfiguration, Cross layer Design.

Download Article PDF