TUJI1 Dataset: Multi-device dataset for indoor localization with high measurement density

Area of Science:

• Computer Science
• Electrical Engineering
• Signal Processing

Background:

• Indoor positioning is crucial in Global Navigation Satellite System (GNSS)-denied environments.
• Existing wireless fingerprinting datasets face challenges with device heterogeneity and sample density.

Purpose of the Study:

• To introduce TUJI1, a novel, anonymized IEEE 802.11 Wireless LAN (Wi-Fi) fingerprinting dataset.
• To address the limitations of device heterogeneity and sample density in indoor positioning research.

Main Methods:

• Collected Received Signal Strength Indicator (RSSI) measurements from 5 commercial devices in a fine-grained grid.
• Anonymized the dataset and matched Wi-Fi fingerprints with precise coordinates.
• Split the dataset into training and testing subsets for reproducibility.

Main Results:

• TUJI1 provides matched Wi-Fi fingerprints and coordinates from diverse commercial devices.
• The dataset is structured for fair and reproducible indoor positioning research.
• Addresses the critical issue of device heterogeneity in Wi-Fi fingerprinting.

Conclusions:

• TUJI1 dataset facilitates advancements in indoor positioning techniques.
• Enables robust and reproducible research by mitigating device variability.
• Supports the development of accurate positioning solutions in GNSS-obscured areas.