Game theoretical framework for clustering and resource allocation in macro-femtocell networks

Abstract

We address the femtocell clustering together with the resource allocation in macro-femtocell networks. The clustering schemes allow the implementation of distributed approaches that can run locally within each cluster. Nevertheless, several limitations should be addressed for dense femtocell deployment, such as: lack of clustering schemes that encourage femtocells to grant service to public users and to become cluster members while guaranteeing their subscriber satisfaction, inefficient bandwidth usage due to the lack of bandwidth adaptation per tier when the cluster configuration changes, and lack of power control mechanisms to reduce interference. In this paper, we propose a distributed clustering model based on a cooperative game, where femtocells are encouraged to cooperate by forming clusters and rewarded with resources from macrocell. Our solution consists of: a cluster formation based on a coalitional game among femtocells and the macrocell to determine the subcarrier distribution per tier, a base station selection for public users and a resource allocation algorithm using Particle Swarm Optimization. We compare our solution with a centralized clustering approach and our cooperative clustering model using the well-known Weighted Water Filling resource allocation algorithm. Simulation results show that our proposal obtains throughput values similar to the centralized approach, satisfies the service requirements for both types of users and reduces the interference in comparison with the benchmark models.