A method for mobile device positioning using a sensor network of BLE beacons, approximation of the RSSI value and artificial neural networks

The paper considers the development of a method for positioning mobile device using sensor network BLE-beacons, approximation RSSI values and artificial neural networks. aim work is to develop small-scale industrial mechanization equipment building unmanned systems product movement tracking. divided into four main parts: data synthesis, signal filtering, selection BLE beacons, translation distance, multilateration. A simplified Kalman filter proposed filtering input suppress Gaussian noise. description two approaches translating value distance given: an exponential function with coefficient determination 0.6994 feedforward network. comparison results these carried out on several test samples: training one, sample at known (0–50 meters) unknown (60–100 meters). shown perform better in all experiments, except meters), which r.m.s. error higher by 0.02 m 2 than that function, can be neglected. An algorithm based multilateration proposed. Experimental studies developed have does not exceed 0.9 meters 5×5.5 room under monitoring. accuracy experiment 40.9 % higher. are also conducted 58.4×4.5 room, showing more accurate compared similar studies.