Teamwork on the Road: Efficient Collaboration in VANETs with Context-based Grouping

There is an emerging trend nowadays for vehicles to be equipped with an embedded on-board computing unit with communication capabilities to enhance the overall driving experience. Such a system enables vehicle-to-vehicle and vehicle-to-infrastructure communication and provides vehicles with up-to-date route and traffic information. However, as this information is not relevant for all vehicles, context-aware communication is considered to be vital for more intelligent inter-vehicular communication. Traditional network addressing schemes are not well suited because they do not take into account contextual properties such as location, direction, speed, time and information interest for group-based communication in large scale vehicular networks. The conventional network paradigms of multicasting and broadcasting to define groups cause too much overhead. First, there is no way to optimize network traffic based on the contextual characteristics of the nodes. Second, they do not take into account the mobility patterns of the vehicles and the road layouts. We propose a context-based grouping mechanism in which only relevant information is shared among the vehicles in a group. We evaluate our approach by formulating groups based on common spatialtemporal characteristics and shared interests. The simulated experiments show that by inducing our context-based grouping mechanism we can significantly reduce the irrelevant/redundant information flows and the overall network traffic usage.