Interference Mitigation in Next Generation Networks Using Clustering and Intelligence Techniques

This work is intended to analyze an interference mitigation model between Small Cells in a dense scenario by using computational intelligence techniques and study the behavior of the mobile users batteries. This is of great importance in the planning and implementation phases of Small Cells networks, since many parameters must be taken into account, about which little or even no information is initially available. Furthermore, we are concerned with the allocation of equipment in such a way to provide the best performance. We observed a problem related to the overall optimization of the system, in which case the use of Genetic Algorithms (GAs) proved to be very effective. In order to address this problem, we first developed an analytical model, in which we could compare the SINR (Signal-to-Interference-Plus-Noise Ratio) values before and after the application of the clustering model and, later, in order to validate the model, we executed simulations and evaluated the Quality of Service (QoS) parameters. We noticed significant improvements in the SINR, achieving about 80% of the Small Cells and keeping the battery consumption behavior stable. Through the simulations, we observed improvements in the quality of the service offered to the users, such as the reduction in the lag above 33%, as well as a drop in the number of lost packets.