Research On Routing Protocols For Flying Ad-hoc Networks

In recent years,unmanned aerial vehicle(UAV)have provided broad and efficient infrastructure solutions for military,industrial and commercial applications,and helped to transform people’s lives into intelligent and interconnected communities.However,current developments in UAVs’ inherent technology,communication networks,and routing protocols limit the applicability of UAVs in many applications.The flying ad-hoc network(FANET)composed of multiple free-flying UAVs has a highly dynamic topology and the energy limitations of UAVs make the design of routing protocols challenging for researchers as they need to consider multiple characteristics such as node speed and energy consumption.Although some improved routing protocols are available,there is still a lack of efficient and suitable routing protocols for UAV networks.Therefore,it is necessary to conduct targeted research on routing protocols to meet the communication requirements in FANET.For the characteristics of fast speed,energy limitation of UAVs in FANET,this thesis proposes a Q-Learning based on-demand flat routing protocol for network topology and link quality awareness.To adapt to the high dynamics of FANET and ensure the packet transmission success rate and delay performance,the protocol applies reinforcement learning technology to network routing,and establishes a link quality evaluation model and an adaptive hello message interval mechanism based on network state information.Therefore,the protocol realizes adaptive adjustment of routing strategy according to network topology and link quality.Simulation results show that the protocol outperforms existing on-demand routing protocols in terms of packet delivery rate,network survival time,and other metrics.In large-scale and high-density FANET,the routing flooding mechanism of the flat routing protocol can cause the collision of network message,resulting in the decline of network communication performance.In view of this,this thesis proposes a hierarchical routing protocol based on fuzzy logic.The main problems of clustering in FANET are the high mobility of UAVs and frequent topology changes.Therefore,this protocol firstly considers the problem of high mobility of UAVs,and will select candidate cluster heads based on node mobility to avoid the participation of highly mobile nodes in the election of cluster heads.Secondly,the optimal cluster head is selected by fuzzy logic considering energy,bandwidth,distance between nodes to complete network clustering.Finally,the communication between nodes is completed by reasonable cluster maintenance and ondemand routing mechanism.The simulation results show that the protocol can effectively improve the packet transmission success rate and reduce the network routing flooding overhead in the large-scale FANET.