Directed graphical game based joint channel selection and power control in cognitive radio networks

Abstract

This article investigates the problem of achieving global optimization for joint channel selection and power control in the cognitive radio networks. Existing schemes modeled the local interference relationship in channel selection by undirected graphs, which however, failed to characterize the directed interference property. In this paper, we reveal the directed and local interference relationship among neighbor cells and propose a directed graphical game to study the distributed channel and power selection problem. Specifically, each transmitter-receiver pair is abstracted as a network access point, that is, a player in the proposed game, among which the interference is abstracted as a directed edge. Then, we prove that the game possesses at least one pure strategy Nash equilibrium. Finally, we propose a joint channel and power selection algorithm, which does not need a controller, to achieve Nash equilibriums. According to the simulation results, the proposed algorithm outperforms the existing schemes in undirected game formulations.