Research On Routing Algorithms In Vehicular Ad Hoc Network

In recent years,with the rapid development of wireless technology,the Vehicular Ad Hoc Network(VANET)has become a research hotspot in the field of Intelligent Transportation.VANET has important applications in traffic accident warning,assisted driving and traffic information acquisition.The routing algorithm is a research focus in the VANET,which can affect the data communication performance of the network.In the VANET,due to the uneven distribution of vehicle traffic and the rapid movement of nodes,the network topology is easily changed and the communication links are frequently disconnected,causing problems such as degraded message delivery performance and high transmission delay.In order to solve the problems faced by VANET,combined with the intermittent connectivity of the Delay Tolerant Network(DTN),designing routing algorithms for the VANET environment has become an important and worthwhile work.Aimed at the problem that the existing algorithms have poor performance in the VANET environment,this thesis will study and improve from the following aspects.The main contents are as follows.(1)For the Probabilistic Routing Protocol using History of Encounters and Transitivity(Prophet)algorithm,there are problems such as large network overhead and low success rate of message delivery,and an improved routing algorithm based on fuzzy logic named FLBR is proposed.In FLBR algorithm,taking into account the nodes' message delivery probability,free buffer ratio,and nodes' movement similarity,the fuzzy logic method is used to obtain the node priority and it is used for routing decisions.At the same time,different message forwarding methods are designed at the intersection and straight road.According to many experiments on ONE simulation platform show that,compared with the Prophet algorithm and other comparison algorithms,the improved FLBR algorithm can significantly reduce the network overhead and improve the message delivery success rate.(2)For the Spray and Wait algorithm based on finite copy control,the network delay is high during message spraying and message forwarding phase.A routing algorithm based on position predicion named PPBR is proposed.In PPBR algorithm,Lagrange polynomial interpolation method is used to establish the position prediction model,and the node's utility value is calculated based on the node's motion information.Then the message spraying and message forwarding are completed by the utility value.According to many experiments on ONE simulation platform show that,compared with the Spray and Wait algorithm and other comparison algorithms,the improved PPBR algorithm can effectively solve the problem of high network delay when the message copys are limited.It can also increase the success rate of message delivery and reduce the average number of message forwarding hops.(3)For the Bubble Rap algorithm based on community divison,when computing the centrality of the node,it needs to obtain the global topology information,resulting in low computing efficiency and affecting message delivery performance.This thesis proposes a distributed community division based routing algorithm named DCBR.In DCBR algorithm,considering the factors such as contact frequency and contact time between nodes,the concept of “social closeness” is proposed.Then the node centrality ranking is calculated according to the node's social closeness,and the message forwarding strategy is designed.At the same time,to reduce redundant messages,a message copy control strategy is also designed in the improved algorithm.According to the ONE simulation platform,this thesis completes many experiments with the external datasets and simulated maps.The results show that compared with Bubble Rap algorithm and similar algorithms,the DCBR algorithm has been effectively improved in message delivery success rate.Moreover,the network overhead and network delay have also been effectively reduced,and the algorithm can be more significantly applied to the VANET environment.