Estimating pedestrian walking time on campus based on wi-fi detection data

Pedestrian travel time is important to the planning and design of pedestrian facilities particularly in high density populated urban areas. With the increasing use of portable electronic devices, the Wi-Fi detection data becomes a promising data source to estimate pedestrian activity patterns. The Media Access Control (MAC) address is a unique signature for each electronic device. In this study, we would make use of these Wi-Fi detection data to extract the pedestrian walking time of crossing a pedestrian tunnel that connects the Phase 8 building to the main campus of the Hong Kong Polytechnic University (PolyU). A data filtering framework is proposed to filter out noisy detections so as to extract the relevant Wi-Fi data. It follows with an efficient solution algorithm to estimate the pedestrian walking time from multiple detection records. Both the means and the variations of walking time are analyzed. The temporal characteristics of pedestrian flow patterns are discussed.