Dynamic cell selection and resource allocation in cognitive small cell networks

Abstract

We consider a sensing-based power allocation scheme in a cognitive small cell network to maximize the sum rate of each small cell by jointly optimizing both the cell selection, the sensing operation and the power allocation over channels, under the condition of interference to primary users below a certain value. Due to intercell interference and the integer nature of the cell selection, the resulting optimization problems lead to a non-convex integer programming which is NP-hard. In order to deal with the non-convexity, we reformulate the problem to a non-convex power allocation game and use the relaxed equilibria concept, namely, quasi-Nash equilibrium. A sensing-based power allocation optimization algorithm that converges to a quasi-Nash equilibrium is also discussed in this paper. Simulation results show that the proposed approach achieves substantial performance gains with respect to a deterministic approach.