Frédéric Le Mouël's Research and Teaching Web Pages: Positions

Open Positions

PhD Proposal

Title

Service Deployment and Invocation under Delay Constraints in Vehicular Networks

Partnership

CIFRE Thesis granted by Valeo Group France in collaboration with the INRIA AMAZONE Team, CITI Lab, INSA Lyon, France.
http://www.valeo.com/
http://www.insa-lyon.fr/
http://www.citi.insa-lyon.fr/

Location

Paris/Lyon, France

Starting Date

January 2012

Contacts

Frédéric Le Mouël (frederic.le-mouel@insa-lyon.fr), Associate Professor
Stéphane Frénot (stephane.frenot@insa-lyon.fr), Professor

Keywords

Internet of Things, Service-Oriented Middleware, Mobile Environments, Vehicular Networks, Disrupted-Tolerant Networks, Autonomous Computing, Decision Algorithm

Context

Internet of Things is populated with a large amount of diverse devices, such as sensors, smartphones, PDAs, home-gateways, etc. All these devices are mobile, intermittently connected and -at the same time- uses and provides services. In vehicular environments, several devices close to the vehicle can host interesting services such as smartphones with the user contactbook providing contact address destinations ; or road signs providing informations to road users about rules, danger, etc ; other vehicles that can provide traffic information about their recent route, etc. Deployment and invocation of these services can not be realized in the same way since some services required a timely-limited access for an immediate availability to the driver, while other services can suffer from some delays.

Challenges

This thesis aims at tackling three issues on service deployment and invocation in vehicular networks:

to express needs and requirements of new vehicular services, such real-time constraints or delay-tolerance,
to integrate pedestrian and vehicular mobility models to estimate possible contact time and anticipate the service exploitation,
to implement the service deployment, discovery and invocation with a dynamic autonomous adaptation to real traffic conditions.

Goals and expected results

In the CITI Lab, we have developed plate-forms [1] providing service deployment, discovery and invocation in a user-oriented ambient environment [2,3].

With on-going works on mobility [4,5], we aim at integrating dynamic service deployment and invocation under delay constraints in vehicular networks in close cooperation with Valeo company.

Thesis expected results are:

a classification of IoT services potentially interesting vehicular networks. This classification should consider the vehicle as an extension of the user or of a group of users and examine their extensions to social networks, collaborative networks, etc.
a domain-specific language allowing to express service availability, with different levels and characteristics such hard real-time, soft real-time, infinite delays. Several languages exist to detail real-time constraints but each time they require to change programming paradigms. A bottom-up approach - based on simple programming paradigms such as Java(-RT)/OSGi - has to be defined with a time data-flow.
an autonomous algorithm anticipating service invocation according to vehicle mobility. As this problem is NP-complete and particularly hard because of the network dynamism, different heuristics have to be developed to meet and balance between hard/soft real-time constraints and user quality of service/experience.
an OSGi implementation integrating a real-time protocol stack and a DTN stack with an automatic service redirection according to service type/constraints.

Application

The candidate should have a strong background in Operating Systems, Middleware, Programming Languages and/or Vehicular Networks. The candidate should submit a complete curriculum vitae and the name of two persons that recommend him or her. The candidate must be fluent in English, French is a bonus. Applications have to be sent before end of November to

Frédéric Le Mouël (frederic.le-mouel@insa-lyon.fr),
Stéphane Frénot (stephane.frenot@insa-lyon.fr),

Bibliography

[1] S. Frénot, F. Le Mouël, J. Ponge, G. Salagnac. Various Extensions for the Ambient OSGi Framework. International Journal of Adaptive, Resilient and Autonomic Systems, Vol. 2, No. 3, pp. 1-12, July-September 2011.

[2] S. Frénot, N. Ibrahim, F. Le Mouël, A. Ben Hamida, J. Ponge, M. Chantrel, and D. Beras. ROCS: a Remotely Provisioned OSGi Framework for Ambient Systems. In Proceedings of the 12th IEEE/IFIP Network Operations and Management Symposium (NOMS'2010), pp. 503–510, Osaka, Japan, April 2010.

[3] A. Ben Hamida, F. Le Mouël, S. Frénot, and M. Ben Ahmed. Déploiement adaptatif d'applications orientées services sur environnements contraints. Computer Science and Technology / Technique et Science Informatiques (TSI), 30(1):59–91, 2011.

[4] A. Tounetki. Ambient Aggregation of Service Repositories. Master Thesis, 2009.

[5] F. Baver Elhuseyni. A Cross-layer Approach to Achieve Efficient Service Delivery in Mobile Delay-Tolerant Networks with a Service-Oriented Network Coding. Master Thesis, 2011.

[6] M. Fiore, C. Casetti, C.-F. Chiasserini, Information Sharing in VANETs, Advances in Vehicular Ad-Hoc Networks: Developments and Challenges, M. Wafta (Editor), Information Science Reference, 2010.

le-mouel.net

Menu: