Distributed learning game based spectrum sharing and resource selection for femtocells

This work investigates enabling technology for spectrum sharing in heterogeneous networks (HetNets) deployment, particularly, when a layer of femtocells (FCs) overlaid upon a mobile cellular network. We propose a fully distributed strategic learning based model that enables Femtocells to autonomously identify spectrum use pattern, and accordingly select available resources, such as to operate under restrictions of avoiding interference and satisfy a certain QoS requirements. The simulation results show that the proposed model can identify unused spectral resources of underlying macrocells network, and FCs can autonomously adjust their spectrum resources and converge to a solution concept that satisfy both networks conditions. We show that intra/inter-tier interference can be reduced significantly, thus resulting in higher cell throughputs. Such a distributed intelligent scheme can provide a practical solution to the main challenges in opportunistic spectrum use and interference management in HetNets.