Indoor Positioning and Tracking by Multi-Point Observations of BLE Beacon Signal

This paper proposes an indoor positioning and tracking method with the aid of large-scale fingerprint observed by received signal strength indicator (RSSI) of Bluetooth low energy (BLE). The RSSI-based positioning is roughly classified into two types: user navigation and monitoring. Our target is the user monitoring scenario by using commercially available BLE dongles, which are compact, inexpensive, and low power consumptions. In this case, the user device sends beacon signals to a large number of receivers in the room. Due to the insufficient BLE transceiver for the accurate positioning, sophisticated support vector machine (SVM)-aided machine learning algorithm plays a vital role in dealing with impairments induced by BLE. Furthermore, according to contiguous movements of the object, prior information of the positioning can be yielded for improving the object tracking capability. Finally, Kalman filter is applied to smooth the tracking results. Experimental results demonstrate the validity of the proposed method.