Distances for wifi based topological indoor mapping

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.

Topic space trajectories

“…Here, we use the Euclidean distance, i.e., dðx; yÞ:¼jjx À yjj 2 . Note that in theory a distance measure for compositional data such as the Aitchison distance (Martín-Fernández et al 1998) or one between probability distributions, e.g., those presented in Schaefermeier et al (2019), would be more suitable for our data. In our investigation, however, we found that the Euclidean distance leads to a better separation of venues into three different research fields, namely neural networks, information retrieval and general machine learning.…”

Topological Indoor Mapping through WiFi Signals

“…Furthermore it provides a smooth distance estimate and hence robustness against random differences in the sampled RSSI observations. It has been shown in previous work [23], that between distributions separated by a large gap, EMD approximates the differences between their means. For overlapping distributions, on the other hand, it captures more subtle differences, e.g., between their variances and kurtoses.…”

“…These steps can be seen as conceptual building blocks, which can be realized in different manners. The steps motion mode segmentation, likelihood estimation and distance calculation have been introduced and evaluated in previous work [23]. For completeness, we briefly sum up the recommended methods.…”

“…In contrast to this, our aim is that each node represents a meaningful abstract location, such as a room. The problem of creating topological indoor maps in our sense was introduced by Schaefermeier et al [23]. In that work, research was focussed on the first steps, namely location representations and distance measures.…”

QoS-oriented capacity planning for edge computing

“…A landmark may be e.g. a payment terminal with a known location [18], or a room or any other indoor structure within a building [121]. Identifying and using landmarks as anchors helps in aligning collected dead-reckoning-based or radio-signal-based traces.…”