Large scale movement analysis from WiFi based location data

Abstract

Understanding and modeling the way humans move in urban contexts is beneficial for many applications. The recent advances on positioning technologies, namely those based on the ubiquity of wireless networks, is facilitating the observation of people for human motion analysis. In this paper we present the result of a large scale work conducted to study the human mobility in a University’s campuses. The study was conducted along several months, using data collected from thousands of users that freely moved inside the numerous buildings existent in two University campuses and a few other buildings in the city center. A Wi-Fi infrastructure of more than 550 access points provides Internet access to the academic community. We tracked the user movements by logging the devices connected to each access point. Based on that data, an analysis process that highlights the relationships between space features and human motion has been developed. In this paper we introduce the concepts of “place connectivity” and “flow across a boundary” to model these relationships. Results show the mobility patterns detected, which are the attraction places along the day, and what places are more strongly connected. This paper also includes an analysis of the short and long term movements between places. With this study we extended our understanding of the life in the campus, enabling us to feel the campus “pulse”.