CaSSaM: Context-aware system for safety messages dissemination in VANETs

Abstract

Context-aware systems have a high potential of application in mobile networks because the context in which they operate is highly dynamic. In particular, vehicular adhoc networks (VANETs) provide scenarios where context-aware systems could be critical to enhance the performance of protocols that depend on network and traffic conditions, to detect hostile environments as well as to offer a novel way to make decisions in real-time. In this paper, we present CaSSaM, a contextaware system that combines information in a decentralized way, from the state of the communications network and the vehicular traffic, with the aim at classifying the scenario in which the VANET is operating. With information about the operation scenario, it will be possible for protocols, for example, a dissemination mechanism, to adequate its parameters with the values that work best in such a context, improving in this way the general performance of the protocol. Our initial results show the evaluation of a well-known dissemination mechanism, namely the ”Slotted 1-persistence”, with different values of parameters according to different operation scenarios. We show how the VANET performance could be improved, or worsened, by choosing different parameter values, and how CaSSaM can help in selecting the proper set of values when the scenario of operation is known.