Article Preview
TopIntroduction
The long distances separating homes and workplaces/facilities/schools as well as the traffic present in these distances make people spend a significant amount of time in vehicles. Thus, it is important to offer drivers and passengers ease of driving, convenience, efficiency and safety. This has led to the emerging of Vehicular Ad hoc Networks (VANETs), where vehicles are able to communicate and share important information among them. VANETs are a relevant component of Intelligent Transportation System (ITS), which offers more safety and better transportation.
VANETs are capable to offer numerous services such as road safety, enhanced traffic management, as well as travel convenience and comfort. To achieve road safety, emergency messages must be transmitted in real-time, which stands also for the actions that should be taken accordingly in order to avoid potential accidents. Thus, it is important for the vehicles to always have available connections to infrastructure and to other vehicles on the road (Kumar & Dahiya, 2020; Sakuraba, Shibata, Sato, & Uchida, 2020). On the other hand, traffic efficiency is achieved by managing traffic dynamically according to the situation and by avoiding congested roads, whereas comfort is attained by providing in-car infotainment services.
The advances in vehicle technology have made possible for the vehicles to be equipped with various forms of smart cameras and sensors, wireless communication modules, storage and computational resources. While more and more of these smart cameras and sensors are incorporated in vehicles, massive amounts of data are generated from monitoring the on-road and in-board status. This exponential growth of generated vehicular data, together with the boost of the number of vehicles and the increasing data demands from in-vehicle users, has led to a tremendous amount of data in VANETs (Xu, et al., 2018). Moreover, applications like autonomous driving require even more storage capacity and complex computational capability. As a result, traditional VANETs face huge challenges in meeting such essential demands of the ever-increasing advancement of VANETs.
The integration of Cloud-Fog-Edge Computing in VANETs is the solution to handle complex computation, provide mobility support and low latency. Each of them serves different functions, but also complements each other in order to enhance the performance of VANETs. Even though the integration of Cloud, Fog and Edge Computing in VANETs solves significant challenges, this architecture lacks mechanisms needed for resource and connectivity management because the network is controlled in a decentralized manner. The prospective solution to solve these problems is the augmentation of Software Defined Networking (SDN) in this architecture.
The SDN is a promising choice in managing complex networks with minimal cost and providing optimal resource utilization. SDN offers a global knowledge of the network with a programmable architecture which simplifies network management in such extremely complicated and dynamic environments like VANETs (Truong, Lee, & Ghamri-Doudane, 2015). In addition, it will increase flexibility and programmability in the network by simplifying the development and deployment of new protocols and by bringing awareness into the system, so that it can adapt to changing conditions and requirements, i.e., emergency services (Ku, et al., 2014). This awareness allows SDN-VANET to make better decisions based on the combined information from multiple sources, not just individual perception from each node.
In a previous work (Qafzezi, Bylykbashi, Ikeda, Matsuo, & Barolli, 2020), we have proposed an intelligent approach to manage the Cloud-Fog-Edge resources in SDN-VANETs using fuzzy logic. We have presented a cloud-fog-edge layered architecture which is coordinated by an intelligent system that decides the appropriate resources to be used by a particular vehicle (hereafter will be referred as the vehicle) in need of additional computing resources. The main objective is to achieve a better management of these resources. In this work, we focus only on the edge layer resources for which we implement two models of our proposed fuzzy-based system that consider the available resources of the neighbor vehicles, among other parameters, to determine their processing capability.