A Fuzzy Logic Reinforcement Learning-Based Routing Algorithm For Flying Ad Hoc Networks

Chenguang He, Suning Liu, Shuai Han

20202020 International Conference on Computing, Networking and Communications (ICNC)31 citationsDOI

Abstract

With the development of technology, unmanned aerial vehicles (UAV) are getting closer to peoples life. Multiple UAV nodes form the Flying Ad Hoc Network (FANET). Due to the high mobility of the UAV nodes, the topology of the flight ad hoc network also changes rapidly. Aiming at the problem of the high average number of hops and low link connectivity, this paper adopts a fuzzy logic reinforcement learning-based routing algorithm for flying ad hoc networks. The fuzzy logic mainly determines the neighbor nodes of a node in real time. Reinforcement learning reduces the average number of hops of the route determined by fuzzy logic through continuous training. Compared with the ant colony algorithm optimization(ACO), the proposed FANET routing algorithm has significant improvement in both link success rate and average hop count. The situation is more perfect and it can better meet the requirements of the network.

Topics & Concepts

Computer scienceWireless ad hoc networkReinforcement learningFuzzy logicOptimized Link State Routing ProtocolComputer networkDestination-Sequenced Distance Vector routingNode (physics)Routing (electronic design automation)Ant colony optimization algorithmsRouting protocolMobile ad hoc networkVehicular ad hoc networkDistributed computingArtificial intelligenceDynamic Source RoutingNetwork packetWirelessEngineeringTelecommunicationsStructural engineeringUAV Applications and OptimizationMobile Ad Hoc NetworksOpportunistic and Delay-Tolerant Networks