A Research On The Cluster Coordinator Selected Mechanism Based On Autonomous Vehicular Cloud

With the development of wireless communication technology and the advancements of functions and qualities of vehicle equipment, the definition of Autonomous Vehicular Cloud(AVC) has been proposed. AVC combines the conventional Vehicular Ad-hoc Networks(VANET) with cloud computing. Due to the mobility of vehicle sources and the dynamic of VANET, aggregation and division of vehicle sources is different from common computer sources. Meanwhile, the process of building AVC has differences with common cloud computing. Therefore, based on the features of AVC, this thesis proposes propose the proper coordinator selected mechanism, and to realize the work of cluster controllers in cloud computing is significant for the progress of AVC.This thesis analyses research status of AVC, combines major technology features both cloud computing and VANET, and proposes cluster coordinator selected mechanism to realize aggregation and division of vehicle sources in AVC circumstance and to achieve the goal of managing and dividing computer sources just like cluster controllers in cloud computing. Cluster coordinator selected mechanism of AVC has two parts:(1) It makes sure coordinator selected criterion depending on computing capacity, memory capacity, network capacity and surviving capacity. It selects the coordinator effectively through the criterion;(2) According to coordinator selected mechanism, it does network clustering locally. The coordinator is to aggregate resources between clusters.This thesis uses OPNET to do theoretical analysis and experimental verification, and evaluate qualities of cluster coordinator selected mechanism. The experimental results indicate that the protocol proposed by this thesis can achieve processing and updating of data quickly and guarantee the validity of local clustering. Through comparison with unstructured resource discovery process, it validates the advantages of the protocol proposed by this thesis.