Context-Adaptive Information Dissemination in VANETs - Maximizing the Global Benefit

Local danger warning (LDW) is one of the most promising active safety applications in vehicular ad-hoc networks (VANETs). Vehicles exchange information about the current road condition and dangerous situations by using WLAN and are able to warn their drivers about upcoming dangers. The information dissemination in particular sets up many requirements that are not entirely fulfilled
by existing concepts. In this paper, a novel concept for a self-organizing and context-adaptive dissemination framework that is capable of combining the advantages of existent routing, broadcasting and dissemination mechanisms is proposed. Message selection and medium access are based on a relevance function that makes use of the current context and the content of the messages. Thus, at any time and in any situation, the message adding the highest global benefit to the system is disseminated. This means
a paradigm shift in the efficient usage of bandwidth because
available bandwidth is consumed as long as there is an increase in the global benefit. The main advantages of the proposed concept are high scalability, an optimal situation and network adaptability, implicit congestion control, and the autonomous formation of dissemination areas. The feasibility and the advantages of the proposed concept are shown by simulation studies which also assist in developing optimized relevance functions.