PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Modeling Energy Constrained Routing in Selfish Ad Hoc Networks
Petteri Nurmi
In: Workshop on Game Theory for Communications and Networks (GameNets'06), 14 Oct 2006, Pisa, Italy.

Abstract

In static networks, game theory has long been used to model the routing decisions of network nodes. However, once we move to dynamic and resource constrained settings, such as ad hoc or sensor networks, traditional models are no longer sufficient. Instead, new models that capture the dynamic nature of the decisions and the resource constraints of the devices are needed. To date, several models that attempt to capture the dynamic nature of routing decisions have been proposed. However, the resource constraints of the devices and the uncertainty about the resources of other devices have been thus far ignored. To this end, we present a game theoretic model that formalizes the resources of the nodes and the beliefs the nodes have about the resources of other nodes. We also discuss the structure of strategies in the proposed model and make explicit the role the resources and beliefs of the nodes play in routing decisions. In addition to presenting a game theoretic model, we propose a method that allows the nodes to learn equilibrium strategies over time, and prove that the strategies suggested by the mechanism converge to a sequential equilibrium. Finally, we present simulations that give insights into the expected behavior of the devices under the proposed model.

PDF - PASCAL Members only - Requires Adobe Acrobat Reader or other PDF viewer.
EPrint Type:Conference or Workshop Item (Paper)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Theory & Algorithms
ID Code:2218
Deposited By:Petteri Nurmi
Deposited On:30 September 2006