Evaluating the Usage of Wi-Fi and Bluetooth Based Sensors for Pedestrian Counting in Urban Areas
Maximilian Franz Böhm, NTNU - The Norwegian University of Science and Technology, Eirin Olaussen Ryeng, NTNU - The Norwegian University of Science and Technology, Torbjørn Haugen, NTNU - The Norwegian University of Science and Technology
In this study, Bluetooth and Wi-Fi sensors are evaluated as tools for counting pedestrians in inner city areas. The measure system is examined in case of its reliability, the further usage and its restrictions.
Evaluating the usage of Wi-Fi and Bluetooth based sensors for pedestrian counting in urban areas
An improvement of the pedestrian infrastructure, as well as a higher amount of people walking in inner city areas are desirable. Pedestrian volumes are one of the key performances to argue for an enhancement of walkability.
Unfortunately, information about the amount of pedestrians, walking in urban areas are regularly based on annual counting’s or traffic model estimations, which are often insufficient for the increasing needs of sustainable inner city infrastructure planning. Automatic measurement systems like infra-red beams or cameras are mostly used and designed for indoor environments like shopping malls or airports. The growing number of mobile devices, equipped with Bluetooth and Wi-Fi interfaces, creates new possibilities in pedestrians’ data collection in indoor and outdoor situations. An automatic, cost effective pedestrian counting approach, operating with Wi-Fi and Bluetooth data to acquire pedestrian information is highly welcomed.
The research question of this study is, whether the usage of Bluetooth and Wi-Fi sensors is suitable for estimating the amount of pedestrians in different inner city areas in a reliable way. The used BlipTrack system is a measurement instrument which detects unique Wi-Fi and Bluetooth Media-Access-Control (MAC) addresses, sent from mobile phones, headsets or other hardware devices.
First, a controlled test is arranged to validate the measurement equipment. In a second phase several field tests are performed and the characteristics of different locations, like bridges, pedestrian zones and arterial main roads are evaluated.
The generated data is analyzed in terms of its penetration rate in different locations and to different times, such as peak an non peak hours. To investigate the penetration rate, a comparison to manual counts is required. The results may show different characteristics depending on the regarded location, and to different times of the day. Filter techniques, like speed ranges or characteristically movement behaviors of the different road users support the distinguishing between pedestrians, cyclists or cars. The measure system is examined and discussed in case of its reliability, the further usage and its restrictions.
Association for European Transport