A leading Mobile Device company needed to harness accurate location detection to enable a vast range of new business cases, such as indoor navigation, inventory tracking, usage efficiency improvement, loss / theft detection, location-based services etc. However, since tracking such high value / critical devices in large indoor premises (warehouses, construction zones, factories) is not reliably possible with regular GPS tracking, alternate connectivity options were required.
Where Wi-Fi coverage was available, a Wi-Fi based solution was developed using a wireless signal strength site map. Where Wi-Fi coverage was not available, BLE beacons were used to make proximity measurements and infer equipment locations. A cloud-based machine learning algorithm was further applied on this data to determine indoor location of devices.
In addition to providing accurate location detection (95-97%), these alternate solutions were also low on capex, which made them easier to deploy on Android and Linux.
Technology Stack:
BLE, Java, Shell scripts, Python, MongoDB, MySQL
