École Nationale Supérieure de
Techniques Avancées

Abstract

The growth of data around the world creates a large number of statistics used to model how individuals live and interact with each other. These interactions can be retrieved, for example, from individual data related to wireless networks. We therefore sought to model these inter-individual networks in order to study their characteristics and try to understand how their structure could be linked to the subjects of virus propagation, social networks and opportunistic networks. To do this, we used GPS, wifi and bluetooth data from the Friends and Family dataset (MIT Human Dynamics Lab, 2011) and used Matlab and RStudio (Statnet and RSienaTest packages) to create these wireless contact networks. A comparison of these networks with other known models allowed us to identify very interesting inter-model correlations linking wireless networks to multimodal mobility and social networks. Finally, this study allows us to understand the relevance of combining different wireless technologies in various situations, whether to obtain a particular graph structure or communication range.