Price-Based Resource Allocation in Spectrum-Sharing OFDMA Femtocell Networks

Abstract

Deployment of femtocells have emerged as a promising technique to address the need for the exponentially increasing mobile traffic demand. They can also improve the capacity and coverage for the indoor wireless users. However, the cross-tier interference in such heterogeneous networks between the femtocells and the macrocells is a design challenge which should be answered before such deployment. In this paper, a spectrum-sharing OFDMA femtocell network is considered in which the macrocell base station protects itself and its users by pricing the interference from femtocell access points. A price-based resource allocation is formulated as a Stackelberg game model to jointly maximize the revenue of the macrocell and the utility of the femtocell users (FUEs) while a predetermined tolerable cross-interference constraint at the macrocell is met. In the proposed model, the macrocell determines the subchannel assignment and the interference prices for the FUEs. Then, the FUEs choose their optimal downlink power. To find the Stackelberg equilibrium, the optimization problems of the leader and the followers are solved. For the followers' sub-game, a closed form expression is obtained. To find the leader's sub-game solution, the non-convex mixed integer nonlinear (MINLP) problem is first converted to an equivalent convex MINLP. Then, we employ the outer approximation (OA) to iteratively and efficiently solve the leader problem. The numerical results validate the convergence of the OA algorithm and the operation of our price-based scheme.