Research On Vehicular Ad-hoc Network Based Cluster And Network Coding

With the rapid development of mobile internet technology,the intelligent transport system incorporating the internet of things,sensor networks and self-organizing networks has been gradually applied in people's daily life.The key component of intelligent transport system is the vehicular ad-hoc network,a mobile self-organizing network composed by communication equipment like on-board unit,road side unit and communication satellite.In recent years,the VANET has attracted great attention of researchers,and becoming one of the major fields of inquiry in wireless communication.Topology diversified and differences of moving direction and speed between vehicle nodes,resulting in the duration time of communication links are extremely short and the quality is difficult to meet the requirements of the message transmission in the vehicular ad-hoc network.In order to solve the above problem,this paper launches the research of VANET technology based on cluster and network coding.Firstly,to improve stability of topology and links in VANET,the using of clustering algorithm is one of important techniques which can also reduce routing overhead.While most existing clustering algorithm of vehicular ad-hoc network are designed for highway environment,which are not suitable for urban scenario where vehicles frequently change their moving directions and speeds.This paper proposes a stable and reliable clustering algorithm based on classical clustering algorithm for VANET.The algorithm considers the optimization of stability condition within urban environment and the influence of channel quality on communication service quality.The result of NS-2 simulation shows that the algorithm could effectively not only improve the stability and reliability of cluster structure but also reduce the transmission latency compared with LID algorithm and MOBIC algorithm.Secondly,a VANET data dissemination strategy based on network coding is proposed.The uneven distribution of vehicle node density leads to the reduction of reliability of data dissemination in sparse scene,so the network coding mechanism is used to improve the reliability of data dissemination.Aiming at the dissemination of map navigation information,random linear network coding is adopted to improve the throughput of the network.In order to reduce the delay of broadcast messages,instantly decodable network coding is adopted for emergency security early warning message system.The simulation results also show that the data dissemination strategy based on network coding has advantages in throughput and delay.Thirdly,a VANET rate allocation algorithm based on congestion control is proposed.There are often multiple vehicle nodes requesting or forwarding messages at the same time,which can easily lead to network congestion.The algorithm takes into account the factors of congestion in the vehicular ad hoc network and the instability of the communication links caused by high speed movement of nodes.To achieve the congestion control and optimal total utility of the network,the optimization method is used to calculate the best transmission rate of the source nodes.Simulation results show that the rate allocation algorithm can effectively control congestion and reduce the influence of link instability on the network.