Reinforcement Learning Based Interference Control Scheme in Heterogeneous Networks

Heterogeneous networks (HetNet) guarantee higher throughput and lower latency than traditional homogeneous network environments. In order to guarantee such performance, interference control between the macro base station (MBS) and the small cell base station (SBS) have to be effectively performed. However, the existing interference control scheme has limitations in controlling interference for rapidly increasing SBS and UEs. In order to solve this problem, we introduce interference control and handover scheme with reinforcement learning in HetNet. Each BS learns transmission power, activation pattern and bias values for optimal network performance in HetNet. We introduce HetNet technologies incorporating various reinforcement learning models and introduce research areas that will be conducted in the future.