On context awareness and social distance in human mobility traces

Abstract

Opportunistic networks consist of mobile devices, carried by people in their everyday lives. They organize autonomously to exchange data with direct neighbors without the use of any infrastructural services. Since the devices are carried by humans, one of the main challenges to consider in opportunistic networks is the human mobility behavior. However, little work exists on how the social behavior of people drives their mobility behavior and how this context information can be systematically leveraged for opportunistic networking applications. This paper tackles this problem by providing both experimental and theoretical analysis of human mobility context information. We present a novel real world experiment with sensor nodes carried by people to demonstrate and study the effect of context on people mobility. Furthermore, we define a novel metric of social distance to put this new evidence on solid mathematical foundation. Thus, our work puts a basis to systematically leveraging context information for opportunistic networking applications and services. Additionally, our experimental data traces enable testing and evaluation of such novel services in a real world scenario.