Research On Adaptive Clustering Algorithm Based On Non-Handshake Mechanism In VANETs

As the number of running vehicles rising up in recent years,the load on roads tends to saturate.Therefore,more and more attention has been drawn to intelligent transportation system(ITS).Vehicular Ad Hoc Networks(VANETs),which is an important part of ITS,has become a research hotspot in the fields of transportation and Internet of Things,both domestically and internationally.Although originated from Mobile Ad Hoc Networks(MANETs),the clustering algorithm proposed for MANETs is not always suitable for VANETs due to the different application scenarios and characteristics.In view of the difficulties of fast moving speed and frequent topological changes of the running vehicles,a framework of adaptive clustering algorithm based on non-handshake mechanism called ACSnH is proposed for VANETs in this thesis.The framework and process of ACSnH clustering algorithm are designed in details to deal with the problems of fast vehicle movement and frequent topological changes in the VANETs scenarios.And then the thesis further focuses on optimizing the process of HELLO packet distribution,cluster head selection and cluster formation on the basis of ACSnH:(1)Aiming at solving the problems of HELLO packet overflow and high communication overhead,an adaptive HELLO packet period adjustment algorithm based on link lifetime called LLTAAP is proposed based on the ACSnH framework.By predicting link lifetime and judging vehicle real-time status,LLTAAP can adaptively adjust its HELLO packet broadcasting period to effectively reduce the communication overhead while maintaining cluster stability.(2)In order to improve the cluster stability and adaptability,a weight-based cluster head selection algorithm called WBCHS is proposed.By considering various traffic factors and selecting cluster heads by weights calculations,WBCHS can effectively improve cluster stability and adaptability.(3)In order to solve the problems of vehicle re-clustering delay and the number control of cluster members,a passive clustering formation algorithm based on range value and non-handshake mechanism called RVnH is designed,and the threshold of cluster head that controls the joining process of vehicles based on adaptive probability is proposed.Byadjusting the joining threshold of cluster head adaptively based on the analysis of potential cluster members,RVnH achieves the goal of controlling the max number of cluster members with non-handshake mechanism,which effectively reduces the vehicle re-clustering delay and improves the cluster stability.Finally,the performance indicators of the proposed ACSnH clustering algorithm are analyzed and evaluated in details.Several different traffic simulation scenarios are built.Coding and relevant simulation experiments of the Lowest-ID,the VMaSC and the ACSnH proposed in this thesis are all completed.The performance of the three algorithms is compared and analyzed.The simulation results show that the cluster stability and clustering efficiency can be effectively improved,and the overhead of cluster maintenance and clustering delay can be reduced by the proposals in this thesis.