Abstract
Vehicular fog computing is an extension of fog computing in the context of vehicular networks. Vehicles are used to serve mobile users at the edge of networks. The vehicular fog is characterized by its dynamicity due to the mobility of vehicles. The change of vehicular fog members over time causes the variation of the available resources. So, the selection of the suitable fog node (i.e. vehicle) to execute a task is challenged. In some cases, the users' requests can be redirected to the cloud to be processed if no suitable vehicle is found to fulfill local processing. To this aim, we suggest a multi-objective based vehicular fog resource allocation scheme. A software-defined networking (SDN) controller is deployed to manage the allocation of the vehicular fog resources based on the information gathered from vehicles and the received requests. The proposed problem is solved using non-dominated sorting genetic algorithm (NSGA-II).