Mitigating Interference in 5G Heterogeneous Networks Through Machine Learning

This paper proposes a machine learning-based algorithm using Reinforcement Learning (RL) for mitigating interference in heterogeneous 5G networks. Via an appropriate design of the reward function, both Macrocell and Femtocell users can access wireless resources more fairly, while maintaining a certain level of total user capacity. Experimental simulation results demonstrate the effectiveness of our approach maintaining user access fairness and improving overall system capacity.