Research And Optimization Of AODV Routing Protocol For Flying Ad-Hoc Network

With the rise of multi-uav cooperative technology in the field of modern digit,Flying Ad-Hoc Network(FANET)has become a research hotspot.Aiming at the scenarios which the nodes move at high speed and the topology change frequently in flying ad-hoc network,the typical Mobile Ad-Hoc Network(MANET)routing protocol is analyzed,focusing on Ad Hoc on-demand Distance Vector Routing(AODV)protocol.In order to improve the stability and applicability of AODV in FANET,this paper concentrates on the following contents:1.An AODV protocol with Polymorphic Awareness(AODV-PA)is proposed to solve the problems of links break frequently and caused by the minimum hop count criterion of traditional AODV protocol and detour caused by routes static update.Firstly,according to the cross-layer design,the link cost based on energy consumption,node stability and buffer congestion is used to select route.Secondly,during the route discovery,the structure of RREQ is modified,and the cache mechanism is replaced by originator sequence number and originator IP address to realize routes dynamic update.Finally,according to the location of source node and destination node,a forwarding strategy,based on fanshaped regions,is proposed to reduce redundant packets.The simulation results of various scenarios and protocols in NS2 shows that the dynamic update can effectively reduce the occurrence of detour,the forwarding strategy,based on fanshaped regions,has obvious effect on restraining flooding,and the improved protocol is more suitable in the topological environment of flying ad-hoc networks.2.An Enhanced Mobility Prediction AODV(EMP-AODV)protocol based on location correction,is proposed to solve the problem of the inaccurate link prediction caused by location errors in UAV Ad hoc network.Firstly,an improved particle filter algorithm is used to compensate the measurement location errors.In the importance sampling process of traditional particle filter,the idea of unscented Kalman is introduced to calculate the mean value and covariance of particles to avoid the particle degradation during the location correction.Secondly,the link expiration time(LET)is predicted by using the rectified node locations,and the route path with the maximum LET is selected.Finally,the route confidence level is considered in the route discovery process,and the link duration confidence is taken as the minimum end-to-end survival threshold to reduce the risky links.Simulation results of NS2 show that compared with the traditional AODV protocol and link prediction protocol without location correction,EMP-AODV protocol improves the overall routing performance,has stronger link stability and robustness,and can better adapt to the highly dynamic topology environment of flying ad-hoc networks.