Research Of Data Dissemination Algorithm Of VANETs Based On Position Prediction And Broadcast Probability In Urban Scenario

With the development of vehicle technology and the popularity of vehicles, VANET(Vehicular Ad hoc Network) has become a new research field of wireless communications. As an important part of the Intelligent Transportation System(ITS), Vehicles carry out data distribution via an self-organization, distributed network environment in VANET, which greatly improves safety and convenience of road traffic. However, since its intrinsic characteristics such as high-speed dynamic topology changes, frequent disconnection, and constantly change of network density, data distribution has become a challenging issue in this area. In this paper, we carry out relevant research of data distribution in urban VANET environment. Firstly, we propose a distributed data distribution algorithm which based on prediction of the node state. This algorithm constructs a set of relay forwarding nodes by predicting speed and position of neighbor nodes. And it achieves data distribution with the way of greedy forwarding according to the density of network which indicates as sparse and dense. This algorithm is mainly used to solve the problem that local neighbor nodes information is outdated and network density changes frequently because of vehicular network topology changes. Secondly, we present a data distribution algorithm based on cluster and broadcast probability. This algorithm builds the corresponding cluster according to the driving direction of vehicle. Cluster head calculates broadcast probability under the local counter and forwarded data according to the calculated probability. This protocol is mainly used for issues of topology changes and frequent disconnection of vehicular network when we distribute data in urban environment.First, we introduce the research background and status of VANET.Second, we focus on the relevant technology in VANET and make a brief introduction of data distribution. In this section, we first introduce architecture and features, practical application and facing challenges of VANET. Then we make a presentation of our existing data distribution protocols in VANET.Next, we propose a distributed data distribution algorithm which based on prediction of the node state to solve the problem that local neighbor nodes information is outdated and network density changes frequently because of vehicular network topology changes. First, we predict the current position and speed of the neighbor node according to its historical beacon information such as speed and location, etc. Followed these measured data, we constructs a prior relay forwarding nodes set; then, we distribute data with the greedy algorithm through the built relay forwarding nodes set considering the change of the network density is sparse or dense.Finally, we present a data distribution algorithm based on cluster and probability to solve the problem of low data transmission rate and large transmission delay caused by topology changes and frequent disconnection of vehicular network. In this approach, we first construct a corresponding cluster in accordance with driving direction of vehicle, meanwhile, the nodes in the network communicate with each other and distribute data through clusters. Local counter of vehicle node conducts the received information statistics and the cluster head distributes data according to forwarded probability calculated by the local counter.