On the performance of vehicular communications with a measurement-based radio propagation model

Abstract

The choice of channel propagation model used for the simulation-based study of Vehicular Ad hoc NETworks (VANETs) plays a key role in the accuracy of conclusions drawn about the performance of applications proposed for these networks. Relying on simplistic channel models, as frequently practised in previous works, leads either to pessimistic or optimistic conclusions. In the view of this, in this paper, we use a realistic measurement-based channel model, implement it in OMNet++ simulator, construct a traffic scenario representing the basic building block of urban traffic networks, and investigate the performance of VANETs in terms of packet reception ratio (PRR). We further compare the performance of VANET under this measurement-based model with two more simplistic channel models used by the practitioners of the existing network simulation tools. Our results show that, for the most part of the simulated scenario, the PRR performance is significantly worse in the measurement-based channel model compared to the PRRs achieved with the simplistic models. However, there is an exception that, in close distances from an intersection and when the sender-receiver pairs are situated on different roads, the measurement-based model leads to better PRR performance.