Privacy-preserving People Counting with Channel State Information

In order to prevent the spread of infectious diseases such as COVID-19 in advance, a method of counting the number of people in the space is needed. However, when the camera is used for counting the number of people, privacy infringement may occur. This method requires an additional cost of installing the camera. With the use of internet of things (IoT) devices and the spread of smartphones, many Wi-Fi access points (APs) are installed. In this paper, we suggest a method to count the number of people in a space using theseWi-Fi APs. First, we show that the wireless channel changes depending on the person in the space. After that, we propose a way to count the number of people through machine learning using channel state information (CSI) of the Wi-Fi signal.