DQR: A Deep Reinforcement Learning-based QoS Routing Protocol in Cognitive Radio Mobile Ad Hoc Networks

In this paper, we propose a novel deep reinforcement learning-based quality-of-service routing (DQR) protocol to establish the best route with minimum end-to-end queuing delay subject to the number of hops constraint in cognitive mobile ad hoc networks (CRAHNs). In forwarding RREQ process, based on the proposed deep reinforcement learning (DRL) model, the DQR protocol unicasts a RREQ packet to its neighbor with minimum cost-value which avoids the affected region of the primary user to save control overheads, queuing delay and routing delay. The simulation results show that the DQR protocol outperforms the AODV one in terms of control overhead, PDR, and delay, suggesting a real-time protocol in CRAHNs.