Abstract
This paper presents a design and an implementation of the Bluetooth local positioning application. Positioning is based on received power levels, which are converted to distance estimates according to a simple propagation model. The extended Kalman filter computes 3D position estimate on the basis of distance estimates. With the used Bluetooth hardware, the mean absolute error of positioning was measured o be 3.76 m. The accuracy can be improved if Bluetooth devices are able to measure received power levels more precisely.