Understanding crowd density with a smartphone sensing system

In this paper, we demonstrate a proof-of-concept prototype of a lightweight indoor crowd monitoring system. The system utilizes off-the-shelf sensors which sniff probe requests periodically polled by people's smartphones in a passive manner. We propose a spatial-temporal data processing algorithm to study crowd density in a given area and their daily routine, based on a passive collection of the probe requests from their smartphones. Moreover, we carry out experiments to show the effect of the transmission interval of probe requests on the network traffic. We also undertake extensive experiments in real-world settings, i.e., one lab room in the university to observe office hours of researchers, and four closely located classrooms on the SUTD University campus to understand room occupancy.