An indoor radio mapping dataset combining 3D point clouds and RSSI.

Area of Science:

• Wireless communication engineering
• Indoor localization and mapping
• Data-driven network modeling

Background:

• Increasing demand for high-bandwidth, low-latency indoor applications (e.g., XR, real-time analytics).
• Need for accurate Radio Environment Maps (REMs) for optimizing wireless network planning and Access Point (AP) placement.
• Limitations of current REM generation methods, including simplified layouts and synthetic data, especially with next-generation Wi-Fi.

Purpose of the Study:

• To address the challenges in generating realistic REMs for indoor environments.
• To introduce a novel dataset for developing and validating REM estimation models.
• To support advancements in data-driven wireless modeling and high-capacity indoor communication networks.

Main Methods:

• Collected a dataset comprising high-resolution 3D LiDAR scans and Wi-Fi Received Signal Strength Indicator (RSSI) measurements.
• Conducted measurements across 20 distinct setups within a multi-room indoor environment.
• Included scenarios with and without human presence to capture environmental dynamics.

Main Results:

• A comprehensive dataset integrating 3D spatial information with radio propagation data was created.
• The dataset enables the study of REM generation incorporating physical geometry and dynamic environmental factors.
• Facilitates the development of more accurate REM estimation models for next-generation Wi-Fi.

Conclusions:

• The presented dataset is a valuable resource for advancing research in data-driven wireless modeling.
• It supports the creation of more robust REMs essential for adaptive indoor wireless network planning.
• Enables the development of high-capacity indoor communication systems by improving radio environment understanding.