Radio Tomographic Imaging (RTI) and Privacy Implications

Radio Tomographic Imaging (RTI) is generally known to be a privacy-preserving localization method. This is due to the reason that RTI does not identify individuals uniquely but only the presence and movements of a human or an object. Most of the existing research on RTI are based on the Zigbee (IEEE 802.15.4) transceivers, but in the present day environments it can be observed that there is a high density of Wi-Fi (IEEE 802.11) networks in public spaces, hence there is a significant improvement of applications for Wi-Fi based localization methods. There are number of Wi-Fi based localization methods in the current literature, but there are limited comprehensive studies on privacy violations that can happen due to the localization methods that are using publicly available data from wireless networks. This research focuses on identifying possibilities of privacy implications that can occur due to Wi-Fi based RTI methods using publicly available Wi-Fi (Received Signal Strength Indicator) RSSI data. In the initial experiments of this research, it has shown that presence of a human in a closed environment can be identified using logistic regression model with a 100% of accuracy level and unique human identification can be done using a multinomial logistic regression model with a 96.3% - 100% accuracy level. These results indicate that there is a possibility of happening privacy implications due to the aforementioned methods.