Deep Learning for QoS-Aware Resource Allocation in Cognitive Radio Networks

This paper focuses on the application of deep learning (DL) to obtain solutions for radio resource allocation problems in cognitive radio networks (CRNs). In the proposed approach, a deep neural network (DNN) as a DL model is proposed which can decide the transmit power without any help from other nodes. The resource allocation policies have been shown in the context of effective capacity theory. The numerical results demonstrate that the proposed model outperforms the scheme in terms of radio resource utilization efficiency. Simulation results also support the effectiveness on the delay guarantee performance.