Realistic and Efficient Channel Modeling for Vehicular Networks (1405.1008v1)

Abstract: To date, VANET research efforts have relied heavily on simulations, due to prohibitive costs of deploying real world testbeds. Existing channel models implemented in discrete-event VANET simulators are by and large simple stochastic radio models, based on the statistical properties of the chosen environment. It was shown that such models are unable to provide satisfactory accuracy for typical VANET scenarios. We performed extensive measurements in different environments (open space, highway, suburban, urban, parking lot) to characterize in detail the impact that vehicles have on communication in terms of received power, packet delivery rate, and effective communication range. Since the impact of vehicles was found to be significant, we developed a model that accounts for vehicles as three-dimensional obstacles and takes into account their impact on the line of sight obstruction, received signal power, packet reception rate, and message reachability. The model is based on the empirically derived vehicle dimensions, accurate vehicle positioning, and realistic mobility patterns. We validate the model against extensive measurements. To enable realistic modeling in urban and suburban environments, we developed a model that incorporates static objects as well. The model requires minimum geographic information: the location and the dimensions of modeled objects (vehicles, buildings, and foliage). We validate the model against measurements and show that it successfully captures both small-scale and large-scale propagation effects in different environments (highway, urban, suburban, open space).