Channel-change games in spectrum-agile wireless networks
Growing numbers of wireless networks such as IEEE 802.11 and Bluetooth are using unlicensed wireless spectrum. Unlicensed spectrum is a shared resource and simultaneous usage by multiple networks can result in interference problems. As the numbers of wireless devices continue to increase, the problem of interference from coexisting networks is expected to become much worse. ^ This thesis introduces modeling of dynamic channel change as a game to address interference from coexisting wireless networks. The emergence of more intelligent, spectrum-agile network components that can dynamically change their transmission characteristics makes dynamic channel change feasible. In contrast, existing work in this area addresses the issue of how to share the current channel more effectively. ^ In this thesis, game-theoretic decision-making based on the self-interest of rational decision makers is applied to a variety of channel-change scenarios. The decision-making consists of selecting the channel-change probability to minimize the transmission delay. Five different scenarios are studied. The results provide a theoretical basis for implementing the decision-making algorithm of a smart access point. ^ Channel-change decisions are studied by first using the simpler approach of single-stage decisions that only consider the present situation. Next, the more realistic but complex approach of multi-stage decisions, that consider expected future actions also, is studied. The "cost" of strictly competitive game-theoretic decisions is determined by comparing them to centralized, socially optimal decisions that maximize the benefit of all coexisting networks. The results indicate that this cost can be considerable. The cost can be lowered by using trust-based schemes such as "coexistence etiquettes". However, in untrusted environments, game-theoretic decisions provide the best outcome. The thesis provides guidance on when to use which approach. ^ The main contribution of this thesis is the introduction of game-theoretic models and their analysis for dynamic channel change decisions in coexisting spectrum-agile wireless networks. The models and analysis have led to an understanding of the policies adopted by self-interest-based decision makers and the resulting system performance. The thesis also provides guidance on when and how to use game-theoretic decisions for channel change. ^
Roli Garg Wendorf,
"Channel-change games in spectrum-agile wireless networks"
(January 1, 2005).
ETD Collection for Pace University.