logo EDITE Sujets de doctorat

Exploiting unmanned aerial vehicles for opportunistically connecting network partitions aiming at efficiently executing distributed applications

Résumé rédigé par
Directeur de thèse:
Doctorant: Joao Paulo DE ARAUJO
Unité de recherche UMR 7606 Laboratoire d'informatique de Paris 6

Projet

Network partitioning in Opportunistic Networks (OppNets) compromises the execution of distributed applications. Providing alternatives to mitigate this lack of communication between disjoint segments is a crucial concern in several scenarios, for example the communication between rescue teams in an emergency situation. This project proposes the use of Unmanned Aerial Vehicles (UAVs) along with an OppNet in order to establish reliable links in disaster areas aiming at efficiently executing distributed applications. UAVs are used to carry physically the messages between network segments by means of a concept known as "message ferrying". The OppNet is hub-based: a hub is defined as a set of nodes in a geographic region. Moreover, besides being a key concept in opportunistic routing, hubs are used for rendezvous between UAVs and ordinary nodes. Methods for exploiting opportunistic routing using hubs combined with UAVs are proposed and analyzed in this project. Thereby, movement models for disaster scenarios and performance metrics will be used. By the end of the project, it is expected a set of new hybrid protocols that exploit both characteristics of the opportunistic communication and the use of controlled-mobility nodes. These protocols must be energetically efficient and fault tolerant, enabling efficient execution of several distributed applications in a network deployed in connectivity-constrained environment. Based on the defined metrics, a comparative analysis with other existing approaches will be also provided.