No-Regret learning for simultaneous power control and channel allocation in cognitive radio networks

Abstract

In this paper, we investigate a no-regret learning algorithm for an exact potential game that allows cognitive radio pairs to update their transmission powers and frequencies simultaneously. We show by simulations that the No-regret algorithm converges to a pure Nash equilibrium, and that it achieves similar performance with the traditional game theoretic framework, while requiring less knowledge about the game and less implementation overhead.