A Q-learning-based downlink resource scheduling method for capacity optimization in LTE femtocells

The deployment of femtocells is beneficial for both users and operators. On the one hand, it can be used to improve the indoor coverage, but on the other hand it will inevitably produce interference issues in the heterogeneous network which consists of femtocells and macrocells. In this paper, a resource scheduling strategy based on Q-learning with Round Robin is proposed. It is compared with conventional scheduling methods from throughput, drop rate, fairness. Simulation results have shown that the proposed method can improve throughput cell-edge users and ensure the requirement of Quality of Service (QoS). It can also implement the compromise of throughput between macrocells and femtocells.