Managing Social Networks in vehicular networks using trust rules

Abstract

Drivers and passengers in urban areas may spend large portion of their time waiting in their cars on the road while commuting to and from work, to school, or to the supermarket. Regularities of driving patterns in time and in space motivate the formation of communities of common backgrounds and interests. We propose a model for forming and maintaining Vehicular Social Networks (VSNs) that uses trust principles for admission to social groups, and controlling the interactions among members. This paper describes the details of the design, and proposes a simple but representative probabilistic model for deriving the probability of wrongful admissions and the probability of an agent trusting a malicious node. The experimental results, which were obtained from simulations using the network simulation software ns2, describe metrics related to the dynamics of group formation and time to form groups as well as to detecting malicious members. Our system was able to form social groups with agents of common interests and maintain an accurate trust evaluation of their behavior.