An adaptive two-phase approach to WiFi location sensing

Environmental variations cause significant fluctuations in WiFi signals in the same location over time, rendering traditional RF-to-location pre-trained maps quickly obsolete. To solve this problem, we use a two-phase approach to determining the user's location. The first phase utilizes traditional pattern-matching to identify the general location, and a second phase applies logistic regression to distinguish between finer-grained locations. An adaptive calibration system allows the user to re-train and dynamically update the signal strength maps to account for the fluctuated signals. We show that our two-phase approach is able to achieve generally high accuracy (-95%) and over in areas of high signal fluctuations due to heavy access point and human density