Joint User-Centric and System-Centric Channel Allocation in Cognitive Radio Networks: A Game-Theoretic Perspective

In this paper, we consider the channel allocation for selfish secondary user (SU) in cognitive radio networks. In general, there are two allocation approaches for our scenario. The user-centric allocation approach aims to maximize the revenues of individual SUs and the system-centric allocation approach attempts to maximize the system revenue. In this paper, we focus on the joint user-centric and system-centric allocation approach and aim to investigate two fundamental issues below: 1) Is there an allocation profile which maximizes the revenues of the SU sand the system? 2) How to achieve this desirable allocation if it exists? We use the game theory approach to investigate these issues. We model the problem of the channel allocation as an ordinal potential game and prove the existence of the desirable allocation profile. Furthermore, we propose a simple protocol to achieve it. Finally, considering the practical issues, we propose a self-enforcing truth-telling method and a best response based algorithm which uses only local information.