BRLR: A Routing Strategy for MANET Based on Reinforcement Learning
Yinghe Wang, Yu Jie Tang · 2021 IEEE 20th International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom) · 2021
Most MANET's topologies have the feature of non-uniform distribution, so the routing protocol based on random distribution topology is no longer applicable. A node betweenness reinforcement learning routing strategy based on network topology control (BRLR) is proposed. In this strategy, parameters such as residual energy and link quality are used to measure the quality of neighbor nodes, and the topology is established. The strategy introduces the node betweenness as a reward factor, and uses reinforcement learning method to establish data forwarding rules. Simulation results show that BRLR has good performance in end-to-end delay, route establishment success rate and so on.