| Vehicular ad hoc network is a new type of mobile self-organizing network that realizes information transmission between vehicles and other communication units.As an important development direction of intelligent transportation,it is widely used in accident warning,traffic control and information service.In the VANET,information distribution is mostly carried out by broadcasting.If the source broadcasting vehicle and the target vehicle cannot communicate directly,the relay vehicle shall be selected to transmit in the way of multi-hop,so as to realize the large-scale transmission of data and information.Therefore,the selection of the relay vehicle is very important in the multi-hop broadcast algorithm.This thesis introduces the key technologies and application classification in VANET,and classifies and describes the existing broadcasting algorithms.In view of the complex network topology and limited messages coverage,the existing multi-hop broadcast algorithms mostly use competition-based relay selection algorithms in different road scenes and select relay vehicles by interacting with local information,it is concluded that these algorithms can easily lead to the inability of timely and reliable transmission of messages to designated locations.In order to improve the overall performance of vehicle communication,a cluster-based network structure is introduced in this thesis.By constructing a virtual backbone network,cluster heads with appropriate number and superior location are selected as relay nodes to assist message forwarding.In the highway scene,a multi-hop broadcast algorithm for the VANET based on global clustering is proposed.In this algorithm,the base station is used to construct the global node topology of the broadcast network.After the vehicles make the broadcast request,the base station calls the fuzzy clustering algorithm to cluster the vehicles in the broadcast network initially.In order to reduce the algorithm complexity,Manhattan distance is used instead of Euclidean distance as a new distance measurement method.Then particle swarm optimization algorithm is used to optimize the cluster center,and relay vehicles are selected according to the node degree of vehicles,the average distance of vehicles in the cluster and channel capacity to assist message forwarding.Compared with the simple single-hop broadcast algorithm,random selection algorithm and K-means clustering algorithm,it is shown that the proposed algorithm can effectively reduce unnecessary broadcast messages,ensure the real-time performance of message transmission and improve the packet reception rate.Aiming at the urban scene with complex wireless communication environment,a multihop broadcast algorithm based on comprehensive weight clustering is proposed.This algorithm comprehensively selects forwarding nodes based on the vehicle’s mobility information,including relative direction,relative distance,relative speed,and relative node degree.In addition,considering the problem of forwarding failure caused by data packet collisions,a timeout retransmission mechanism is designed to ensure the effective transmission of messages.Finally,the SUMO software is used to build a real road scene and vehicle movement model.The simulation results are compared with the flooding broadcast algorithm and the multi-hop broadcast algorithm based on urban road layout,and the performance of the algorithm is verified in terms of message coverage,broadcast delay and the proportion of forwarding nodes. |
