Connected Cars: Mobility and Connectivity challenges.
Sujet proposé par
Directeur de thèse:
Unité de recherche UMR 7606 Laboratoire d'informatique de Paris 6
Domaine: Sciences et technologies de l'information et de la communication
Mobility Challenge: Understanding mobility is key to support services and applications while on the go. Properly modeling urban mobility to factor in user behaviors, and correlations between different users and with activities is essential for the design and evaluation of new communication protocols. We plan to cooperate with the partners to analyze data from various installations and field trials with the aim of characterizing user mobility and exploit its feature in both protocol design and infrastructure design. For example a detailed knowledge of user mobility may help to deliver heavy content such as video to strategic connectivity points along the way such as for instance access points installed on top of a traffic light.
Connectivity Challenge: A large number of advanced services require users to be connected to the Internet. The current model based on cellular operators is however showing its limits in both performance and costs. Operators are struggling to increase their ARPU while reducing their CAPEX therefore the market is shaped by per/byte contracts and relatively small forfeit (i.e. les than a full HD movie per month). A new approach to mobile network is needed. The chair will explore a new generation of communication protocols able to exploit multiple communication channels at the same time and to benefit from in-network caching. This will require a radical change in the protocol architecture and in the communication paradigm. Specifically we will investigate Information Centric Networks for connected vehicles. The ICN paradigm elects as communication endpoints the content rather than the communication node. This change in perspective leads to new network architectures able to take advantage of the best available link on a per-packet basis. The in-network caching, key part of Information Centric Networking is functional to eliminate unneeded wireless transmissions and to reduce the time needed to fetch contents from the network. Applications such as for example video-on-demand, maps update, and predictive maintenance would benefit from caching, and disruption tolerant delivery thus drastically reducing the needs for an always-on 3G connection. At the lower layers of the communication stack, closer to the technology, a fine tuning o the wireless capabilities is needed. The current IEEE8021.11 needs to be optimized and tuned to work in high mobility scenarios such as the V2V and V2I case. Specifically, there are several open challenges:
(a) the design of fast-attach mechanism that would allow mobile devices to benefits from communication opportunities presented by access points available on the road;
(b) channel sensing techniques to optimize the V2V communications towards a reduced packet loss probability and an increased delivery rate.