Noncooperative Equilibrium Solutions for Spectrum Access in Distributed Cognitive Radio Networks

This paper considers the problem of channel selection and dynamic spectrum access in distributed cognitive radio networks. The ability of a cognitive radio to adaptively switch between channels offers tremendous scope to optimize performance. In this paper, the dynamic spectrum access in a distributed network is modeled as a noncooperative game and the equilibrium solutions are obtained through a bimatrix game. The cost term of the utility function and the several possible definitions of "price" and how they characterize the equilibrium solutions provides a new perspective on the analysis. In distributed cognitive radio networks, the secondary users are vulnerable to several unexpected events such as primary user arrival or a deep fade or sudden increase in interference which could potentially disrupt or disconnect the transmission link. In such cases, any strategic decision or information that could lead to uninterrupted channel access and facilitate maintaining links could be modeled as a Stackelberg game. Performance characteristics for both the leader and follower nodes for the defined utility functions are given.