An Intelligent Routing Decision-Making Technique Based on DQN for MANETs

The rapid advancement and utilization of Mobile Ad Hoc Networks(MANETs) have led to the widespread application of deep learning in intelligent routing issues. While current intelligent routing algorithms concentrate on identifying the best next-hop node through deep learning, they overlook the influence of the dynamic external electromagnetic environment and their own network setting on the routing approach. Consequently, this paper introduces a deep reinforcement learning-based intelligent routing decision-making technique. Tailored to diverse routing approaches needed for various service types and intricate link conditions, this method allows nodes to flexibly adapt their routing strategies in response to real-time service requirements and environmental variations. Through simulations, it is observed that the intelligent routing policy decision-making technique exhibits superior overall performance compared to a single routing policy.