ABSTRACT
Vehicular networking is an emerging area of interest in the wireless networking community as well as in the transportation research community. e potential of vehicular networks to provide vital services, from real-time tra c information to advance collision warning, makes this an important area of study. Vehicular networking can comprise vehicle-to-vehicle (V2V) communication, vehicle-to-infrastructure (V2I) communication, or a combination of both. Typically, networks formed without infrastructure support are termed ad hoc networks; thus, vehicular networks with only V2V communication have been called vehicular ad hoc networks (VANETs). VANETs have much in common with the well-studied mobile ad hoc networks (MANETs). Both are ad hoc networks of mobile nodes that are capable of wireless communication. Much of the similarity ends there, however. Because VANET nodes are vehicles rather than handheld devices, there is little concern with energy consumption, storage capacity, or computation power. Additionally, because vehicles move much faster than humans, VANETs
have the added complication of a quickly changing topology. In order to evaluate VANETs*, researchers almost always must resort to simulation as the expense of actual deployment is too high. Unfortunately, there is no standard vehicular networks simulator. Currently, most researchers generate a mobility trace using a vehicular mobility simulator and input this trace to a standard networking simulator. e choice of the mobility simulator is important as performance in vehicular networks depends highly on the connectivity of the nodes, and the manner in which nodes move greatly a ects this connectivity.
