Effect of non-cooperative vehicles on path connectivity in vehicular networks

Abstract

A fully-connected Vehicular Ad-hoc Network (VANET) contributes tremendously to the improvement of the travel experience of commuting passengers. However, the existence of an end-to-end multi-hop path across VANET highly depends on the willingness of vehicles to cooperate with one another when it comes to data forwarding. Unlike the existing VANET connectivity studies that focused solely on fully-cooperative vehicular environments, this paper develops a mathematical model to explain the effect of non-cooperative vehicles on the end-to-end connectivity through the roadway segments of a VANET.

First, this work characterizes some of the fundamental traffic-theoretic properties of VANETs in the presence of non-cooperative vehicles. Then, it proposes to alleviate the detrimental effect of non-cooperation on end-to-end path connectivity by exploiting Unmanned Aerial Vehicles (UAVs) as store-carry-and-forward nodes in a VANET. Through both mathematical analysis and extensive simulations, the role that UAVs can play in enhancing end-to-end path connectivity is quantified in the context of a hybrid, UAV-assisted VANET architecture.