Social-based forwarding in multi-channel vehicular networks

Recently a lot of effort has been dedicated to maximize throughput in highly mobile multi-channel networks such as vehicular networks. Nevertheless, distributing the load as equally as possible between service channels varying in space and time is still a matter of investigation. The contributions of this paper are two-fold. Firstly, we present an adaptive multi-channel allocation mechanism (AMC) that relies on space-time discretization to fairly balance the load on service channels. Secondly, this paper investigates how the delivery performance of service updates can be enhanced by targeting vehicles which exhibit higher centrality. A Social-based Relaying Strategy (SRS) which optimizes the dissemination of service updates to create a wider situation awareness between areas of interest is introduced. This stateless strategy exploits knowledge from the social properties of the communication graph into its forwarding´s weighting function. More particularly the weighting function includes a localized version of the clustering coefficient metric and the lobby index in order to select forwarding vehicles with high centrality and strong connectivity. Simulations based on a post-crash scenario demonstrate that SRS can substantially improve the delivery performance compared to GPSR-MA, a stateless geographical forwarding algorithm and slightly outperforms another social-based forwarding strategy which only accounts for lobby index.