(A) study on experimental model for evaluating applicability of local vehicular dynamicity in connected car environments

Abstract

The connected car market is currently growing and it is expected that the application services which collect, process and utilize the real-time local dynamic information in the cloud will be increased through the technological developments. However, there exists no management model to provide a technical feasibility check of such connected car application services. To manage and evaluate such application services, we proposed a definition of Local Vehicular Dynamicity (LVD) and the LVD application service environments. In addition, we proposed a new method for evaluating the adaptability of LVD, which derives the procedures of LVD applications and the computing processing time according to the response time Service Level Agreement (SLA). With the proposed method, we evaluated the statistical analysis of LVD which is collected and analyzed through the connected car application and the application service execution time estimation in the cloud environment. Furthermore, we experimentally analyzed and evaluated the model which improves the performance by utilizing Probabilistic Inverse Simulation (PIS). Finally, we experimentally implemented and evaluated the emulatedsimulation with BlindMotion as a basic LVD application service. In the future, we expect that the feasibility of future LVD applications will be able to be evaluated with the emulated-simulation whose accuracy will be increased by collecting real data from users.