Zee: Zero-Effort Crowdsourcing for Indoor Localization

While RF fingerprinting based on WiFi or cellular signals has been a popular approach to indoor localization in the research literature, its adoption in the real world has been stymied by the need for site-specific calibration, i.e., the creation of a training data set comprising WiFi measurements at known locations in the space of interest. While efforts have been made to reduce the calibration effort using modeling, the need for measurements from known locations remains a bottleneck. On the other hand, localization based on inertial sensors in a mobile device carried by a user has the advantage of not requiring site-specific calibration, but it suffers from user-specific peculiarities such as the placement of the mobile device and the user’s walking characteristics.

In this paper, we present Zee, which makes calibration zero-effort, in a way that training data can be crowdsourced without any explicit effort on the part of users. The only site-specific input that Zee depends on is a map showing the pathways (e.g., hallways) and barriers (e.g., walls). For the rest, Zee makes measurements using the accelerometer and compass sensors, and performs WiFi scans. It then employs a suite of novel techniques to infer location over time: (a) placement-independent step counting and orientation estimation, (b) augmented particle filtering to simultaneously estimate location and user-specific walk charateristics such as the stride length, (c) back propagation to go back and improve the accuracy of localization in the past, and (d) WiFi-based particle initialization to enable faster convergence. We present an evaluation of Zee in a large office building.