Resource Allocation For D2D Communications With A Novel Distributed Q-Learning Algorithm In Heterogeneous Networks

This study investigates energy efficiency issues of device-to-device (D2D) communications in heterogeneous networks. To minimize the total transmit power, an approach based on Q-learning together with logarithmic cooling (LC) is proposed to optimize the connection of user equipment (UE) with base station (BS) or Access point (AP). The proposed distributed Q-learning (D-Q-learning) with LC can conduct the adequate exploration and exploitation operations for effective optimization. Simulation results indicate that in the multi-cell scenario, the proposed D-Q-learning can obtain the sub-optimal performance close to the best solution.