Decentralized Learning Algorithm for LTE and Femtocells Networks
Femtocells aim to improve the cellular network coverage and capacity. However, new conception challenges are posed by deploying the femtocells randomly on the cellular network. One of the major issues is the interferences caused by the cross-tier transmissions between femtocell and macrocell which both work on the same spectrum. Among all the access control mechanisms, hybrid access seems to be the promising choice, since the femtocellopens a part of its resources for macro users while reserving the residual part to its own users. We propose a hybrid access control mechanism, where the macrocell remunerates a refunding amount to femtocells depending on their contribution to macro user’s data trans- mission. The aim of this work is to study the environment concurrence between the femtocells with a decentralized manner by using the learning algorithm LRI. Some simulations have been conducted and the results show that the utilities of both, femtocell and macrocell, are significantly improved exploiting thehybridaccess mechanism.