Abstract
A major burden of signal strength-based fingerprinting for indoor positioning is the generation and maintenance of a radio map, also known as a fingerprint database. Model-based radio maps are generated much faster than measurement-based radio maps but are generally not accurate enough. This work proposes a method to automatically construct and optimize a model-based radio map. The method is based on unsupervised learning, where random walks, for which the ground truth locations are unknown, serve as input for the optimization, along with a floor plan and a location tracking algorithm. No measurement campaign or site survey, which are labor-intensive and time-consuming, or inertial sensor measurements, which are often not available and consume additional power, are needed for this approach. Experiments in a large office building, covering over 1100 m2, resulted in median accuracies of up to 2.07 m, or a relative improvement of 28.6% with only 15 min of unlabeled training data.
Highlights
Localization and tracking in indoor environments is important for a wide range of location-aware applications, e.g., museum guidance, navigation in a shopping mall, finding your car in a parking garage, or asset tracking in the industrial sector
It uses an initial radio map based on a theoretical path loss model, unlabeled training data, a self-calibration method, and a route mapping filter
The premise of this work is that the differences between real measurements and reference values, derived from a model-based radio map, tend to be correlated per room and access point
Summary
Localization and tracking in indoor environments is important for a wide range of location-aware applications, e.g., museum guidance, navigation in a shopping mall, finding your car in a parking garage, or asset tracking in the industrial sector. Most positioning systems in GPS-denied environments rely on signal strength measurements from existing wireless network infrastructures due to their simplicity and availability, e.g., WiFi, ZigBee, or Bluetooth Low Energy (BLE) compatible devices. These Received Signal Strength (RSS) measurements can be translated to a location by making use of a path loss model and the well-known multilateration method [1]. The fingerprint database, or radio map, is a signal space that links RSS values to positions in a building This database is constructed in an offline phase by making use of a radio channel simulator or an elaborate measurement campaign, known as war-driving [4].
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call