Development and Testing of a Real-Time WiFi-Bluetooth System for Pedestrian Network Monitoring and Data Extrapolation
Asad Lesani, Luis Miranda-Moreno · Transportation Research Board 95th Annual MeetingTransportation Research Board · 2016
Real-time data collection and monitoring of pedestrian networks is an important topic in research and practice. A real-time pedestrian monitoring system should be able give information about volumes (counts or flows) and speeds for all links of interest. There are many applications for pedestrian network monitoring such as counting and extrapolation, activity modeling, management of public areas like airports, train and metro stations, malls. Moreover, there are some more specific applications like security checks, waiting time and travel time in public hubs like airports. There are some available technologies to get count data at specific locations; however these types of technologies cannot provide detailed data of pedestrian walking patterns and speeds. To achieve this, the system should be able to collect anonymous data of pedestrians in a network. In recent years, Bluetooth technology has widely been used in studies to capture the unique Media Access Control (MAC) addresses of Bluetooth devices in order to track them throughout a network to assess their activity patterns. Because of the security and power consumption concerns and less applications due to new wireless protocols like WiFi, the penetration of this technology is getting smaller in smartphones. Hence, an integrated WiFi-Bluetooth system was designed in order to take advantage of the benefits of WiFi technology like higher penetration rates between users of smartphones. Some criterion like travel time and detection rate are used in this paper to evaluate the performance of the developed system. The primary results show high detection rates of WiFi system as well as, a high correlation between the number of detections between sensors and ground truth data. These features can be helpful for extrapolation and estimation of pedestrian flow furthermore more accurate travel time and waiting time estimations.