Assessing BLE RSSI Fingerprinting for Product Tracking in a Warehouse

Indoor tracking of products in a warehouse is a challenge that requires a scalable and reliable solution. Among those challenges are energy efficiency, positioning accuracy, scalability and coverage. However, when using radio-based indoor localization and tracking systems, the major challenge is dealing with the characteristics of the operating area and of the target objects, where the presence of metal objects deeply affects the propagation of radio signals. Fingerprinting based on Bluetooth Low Energy (BLE) Received Signal Strength Indicator (RSSI) is a cheap and straightforward technique. However, it suffers in terms of accuracy compared to other solutions using different technologies and/or measurements to implement an Indoor Positioning System (IPS), offering a trade-off between complexity, cost, and accuracy. Product tracking inside a warehouse may not require sub-meter accuracy, but it requires a scalable and reliable solution. To this end, BLE RSSI Fingerprinting was assessed to assert its viability in these scenarios. This paper describes the implementation of an IPS using BLE RSSI Fingerprinting to track pallets inside a warehouse, which was tested in a real-world setting, achieving an average error of 2.24 meters.