Dynamic Topology Discovery Configuration in Software-Defined Vehicular Networks

The advances in the automotive industry and radio access technologies along with new relevant business opportunities and applications, ranging from road safety to infotainment, bring fresh interest for the Vehicular Ad-hoc Networks (VANETs). However, the conventional architecture of VANETs along with their dynamic nature make deployment of new services/protocols, network management and routing, among others, quite arduous tasks. Software-defined networking (SDN) brings centralized control, programmability and flexibility to vehicular networks, paving the way for a novel networking paradigm termed as Software-Defined Vehicular Networks (SDVNs) that leverages them to satisfy their performance and management requirements. In this article, we present a dynamic topology discovery (TD) approach for SDVNs that allows efficient data collection, through Vehicle-to-Infrastructure (V2I) communication, in terms of packet delivery ratio (PDR) and control overhead. The proposed approach utilizes vehicles’ traffic-related data. It proceeds with their analysis and, then, it specifies individual, on-the-fly configurations allowing each vehicle to achieve connectivity with the nearest fixed infrastructure, e.g., Road Side Unit (RSU). To apply such dynamic configurations, we employ the strategies of SD-MIoT, a novel SDN solution for mobile Internet of Things (IoT). Analytical results guide our simulation which shows that dynamic TD configuration (dTD) brings improvements in the PDR up to 31.5%, while maintaining control overhead at low levels compared with static configuration (sTD).