Distances for WiFi Based Topological Indoor Mapping

TL;DR

This paper evaluates different distance measures for WiFi-based indoor mapping, finding Earth Mover's Distance most effective when combined with kernel density estimation to preserve topological room structures.

Contribution

It compares various likelihood distance measures for WiFi localization and demonstrates Earth Mover's Distance's superiority in preserving topological features.

Findings

Earth Mover's Distance outperforms other measures in localization tasks

Kernel density estimation helps retain topological structure of indoor spaces

Effective WiFi-based topological mapping demonstrated in real-world scenario

Abstract

For localization and mapping of indoor environments through WiFi signals, locations are often represented as likelihoods of the received signal strength indicator. In this work we compare various measures of distance between such likelihoods in combination with different methods for estimation and representation. In particular, we show that among the considered distance measures the Earth Mover's Distance seems the most beneficial for the localization task. Combined with kernel density estimation we were able to retain the topological structure of rooms in a real-world office scenario.