Evaluating Fingerprinting-Based Indoor Localization Using RSSI Data from BLE Devices

Indoor localization is a rapidly evolving field with applications ranging from navigation to asset tracking and intelligent environments. This study explores the potential of fingerprinting-based localization using Bluetooth Low Energy (BLE) devices, leveraging a newly released open access dataset collected in a real-world school environment. The dataset’s rich configuration and comprehensive signal data provide a unique opportunity to evaluate the performance of a straightforward yet effective 1-Nearest Neighbor (1-NN) algorithm. Our findings reveal promising localization accuracy and low error rates, highlighting the practical value of this approach in challenging indoor settings. By presenting baseline performance metrics, this work lays the groundwork for advancing robust, scalable indoor positioning systems that address real-world challenges such as signal attenuation and multipath interference.