Research On Data Broadcasting Strategy For Vehicular Ad Hoc Networks In Urban Scenarios

In order to solve problems such as traffic accidents and road congestion,Intelligent Transportation System(ITS)came into being.The research focus of ITS is Vehicular Ad Hoc Networks(VANETs).The special characteristics of VANETs bring great difficulties to the research,such as the high speed of vehicles,continuous change of network topology,and the communication quality being affected by buildings and surrounding environment greatly.While some characteristics bring benefit to the research such as predictable trajectory of vehicles,accurate location information.The goal of this thesis is to transmit emergency message to vehicles around and further away when an emergency occurs on urban roads,so that other vehicles know the accident as soon as possible and adjust travel routes in time to bypass the accident area.The main contributions include following three parts:(1)The adapting strategy of sending interval for hello message is studied.The hello message is the basis of neighbor discovery,establishment of neighbor table,and information exchange for vehicles.The current sending interval of hello packets is mostly a fixed value,and it consumes a lot of network resources in a vehicle-intensive scenario,which can seriously cause network congestion or paralysis.Therefore,this thesis proposes an Adaptive Hello interval strategy based on Neighbors Changing Rate(AHNCR),which designs to reduces the frequency of hello message in vehicleintensive scenarios for saving network resources.Based on the result of simulation,AHNCR can effectively reduce the overhead while ensuring the accuracy and reliability of neighbor table.(2)The data broadcasting strategy for vehicular ad hoc networks in urban scenarios is studied.In current research,the p-persistent broadcast protocol is with good delivery rate and delay,but usually consider only one factor,such as distance between nodes,the number of neighbors.In realistic scenarios,the distance is related to the number of neighbors and they should be considered simultaneously.Therefore,this thesis proposes a new data broadcasting strategy,P-persistent Broadcast strategy based on Neighbors with Distance(PBND).Both the distance and the number of neighbors will affect the probability in PBND's model.When a node has few neighbors,the node at the same location will get a higher forwarding probability to ensure that the packet can be successfully forwarded.Based on results from a packet-level simulation,the PBND strategy performs well in dense and sparse traffic scenarios.(3)For urban scenarios,the law of traffic volume changing over time is summarized.The traffic simulation software SUMO is used to establish three typical cases of different volume of traffic which obtain from the summarized law,which makes the simulation closer to real situation.The proposed AHNCR and PBND are simulated in the three cases in Veins,and compared with existing broadcasting strategies.Results of simulation show that the proposed method can better adapt to urban scenarios with changing traffic flow.