Whereas passive safety systems more and more reach their physical limits, active safety systems are going to determine the road map towards a "zero-accidents" vision. A big potential is seen in Vehicular Ad Hoc NETworks (VANETs). They are expected to go far beyond the capabilities of local radar- and vision-based sensors, by providing an enhanced view of the current environment, known as cooperative awareness. To this objective, vehicles are compelled to periodically broadcast safety-related information (e.g. position, speed, heading) to their immediate neighbors over a modified IEEE 802.11 technology adapted to the vehicular environment. Despite its flexibility, this technology cannot guarantee reliable vehicular communication, and the transmission losses from packet collisions represent a challenging issue to the reliability and trustworthiness of cooperative traffic safety applications.
In this thesis, we therefore analyze the broadcast collision problematic for cooperative vehicular applications, and address it in in three phases: we first conduct a thorough analysis of the source of packet collisions in VANET. We then propose to mitigate them by adapting the transmit. We finally apply these transmit adaptation strategies to a cooperative vehicular safety applications and prove their feasibility and their efficiency to enhance the reliability of cooperative safety applications.
From the packet collision analysis, we identify that significant collisions are correlated in space and in time due to the specific transmit policies required by cooperative traffic safety communications, leading to a severe reduction of the reliability and the efficiency of cooperative safety applications. Accordingly, one objective of this thesis is to propose transmit adaptation concepts in order to mitigate these correlated packet collisions in space and in time.
Addressing the time aspect first, we propose the concept of random transmit jitters to mitigates correlated packet collisions by randomizing the transmissions around the nominal broadcast interval. On the spatial aspect, we introduce the random transmit power concept, which randomizes packet collisions in space, and avoids the collision area to span over the same vehicles for multiple successive safety messages. We then illustrate the benefit of these transmit adaption strategies on cooperative traffic safety applications. We introduce a framework called fish-eye awareness, which proposes to adapt the required awareness quality as a function of space rather than forcing the VANET to provide equal awareness everywhere in space. We further show how the transmit adaptation strategies manage to fulfill this framework and optimize channel resources in space and in time. Through this thesis, we demonstrate the gained benefit both in terms of communication reliability and in application reliability obtained from the proposed broadcast mitigation strategies.