Estimating Pedestrian Traffic Volume using Wi-Fi Packet Sensor and Signal Strength

Abstract

It is important to understand the pedestrian traffic volume, which is one of the indicators to quantify the liveliness of the city and tourist attractions. In this study, we quantified the difference between the data obtained by the Wi-Fi packet sensor and the actual pedestrian traffic observed, and investigated the data cleaning method using the signal strength. At four locations in Obuse Town, Nagano Prefecture, pedestrian traffic flow was captured with a video camera and the traffic volume was measured for 15 minutes. A sensitivity analysis was conducted to set the threshold of the signal strength. This made it possible to clean the data according to the environment where the sensor was installed. Then, we performed a regression analysis on the number of pedestrians using the Wi-Fi packet sensor. The results showed that the accuracy of the estimation is sufficient for the case where the increase or decrease of the number of pedestrians is required throughout the year.