A Cluster-Based Vehicular Cloud Architecture with Learning-Based Resource Management

Recently, involving wireless communication technologies in deployment of new vehicular networks becomes more attracting to the research community and the vehicle manufacturers. It is beneficial in providing intelligent transportation system as well as new assistant services to drivers. However, the limitation of resources in mobile vehicles is a significant technical challenge in the deployment of new applications. In this paper, we propose a new vehicular cloud architecture used clustering technique to group vehicles and provide resources cooperatively. We make the cluster structure flexible by using the fuzzy logic in the cluster head selection procedure. Also, we improve the resource management of our architecture by employing the Q-learning technique to select a service provider among the participant vehicles. Finally, performance of the proposed architecture is evaluated using extensive simulation and the efficiency of its components are demonstrated through comparison with the other existing approaches.