Title: Context awareness and service discovery in hybrid networks
Author: Tom Leclerc
e-mail: tom.leclerc@loria.fr
Partner: INRIA
Supervisor: André Schaff and Laurent Ciarletta
Committee:
Year of start: 2007
Year of end: 2010
Funding institution: Centre de Recherche INRIA Nancy - Grand Est

Hybrid dynamic networks also known as mesh networks are a particular case of wireless ad-hoc networks in which one part of the nodes stays fixed and the other one is strongly dynamic. The service discovery protocols (e.g. Jini, UPnP and Salutation) currently available enable access, in a transparent manner, to the services provided on the network (e.g. printing service or Internet access). However, these protocols are not adapted to mesh networks because of the latter’s strong dynamicity and heterogeneity in terms of hardware, software, connectivity and usage. The context (cartography, localization, situation etc.) in dynamic networks is still not taken into account. This lack of context not only concerns service discovery protocols, but also most routing protocols, including those dedicated to dynamic networks. This thesis will mainly focus on studuying and proposing metrics that take into account the context in order to improve both service discovery and routing protocols in MANETs (Mobile Ad hoc NETworks) and in Mesh networks, using the peculiarities of these networks at our advantage. It goes from basic metrics like power, transmission rate to more specific metrics focused on service discovery. The goal is to combine these metrics, by adding weights, into a single metric.

In order to evaluate and validate these metrics we will on one hand use and develop simulation tools that can simulate complex mobility and protocols, and on the other hand we will evaluate our solutions in a real world experiment that will be conducted in the frame of the ANR - SARAH project, most probably a “museum of new technologies”.