Deep Q-Learning Framework for Improving Spectral Efficiency in D2D Communication

Device-to-device (D2D) communication is expected to play a big role in 5G, in order to enable new applications in the mobile system such as vehicular-communications (V2X) and internet of things (IoT). This should increase the bandwidth demand, which makes higher spectral efficiency ever more desirable. This paper proposes a Deep Q-Learning power allocation framework for maximizing spectral efficiency in D2D communication, in underlay mode, while satisfying the mobile user's quality of service (QoS) requirements.