XJTLUIndoorLoc: A New Fingerprinting Database for Indoor Localization and Trajectory Estimation Based on Wi-Fi RSS and Geomagnetic Field

TL;DR

This paper introduces a new Wi-Fi and geomagnetic fingerprinting database for indoor localization and trajectory estimation, demonstrating the effectiveness of CNN and LSTM models trained on this data.

Contribution

It provides a novel multi-floor indoor fingerprinting database and preliminary deep learning-based methods for localization and trajectory estimation.

Findings

Database proves useful for indoor localization tasks.

CNN effectively maps Wi-Fi RSS data for position estimation.

LSTM accurately models human walking trajectories.

Abstract

In this paper, we present a new location fingerprinting database comprised of Wi-Fi received signal strength (RSS) and geomagnetic field intensity measured with multiple devices at a multi-floor building in Xi'an Jiatong-Liverpool University, Suzhou, China. We also provide preliminary results of localization and trajectory estimation based on convolutional neural network (CNN) and long short-term memory (LSTM) network with this database. For localization, we map RSS data for a reference point to an image-like, two-dimensional array and then apply CNN which is popular in image and video analysis and recognition. For trajectory estimation, we use a modified random way point model to efficiently generate continuous step traces imitating human walking and train a stacked two-layer LSTM network with the generated data to remember the changing pattern of geomagnetic field intensity against (x,y) coordinates. Experimental results demonstrate the usefulness of our new database and the feasibility of the CNN and LSTM-based localization and trajectory estimation with the database.