A WKNN-based approach for NB-IoT sensors localization

With the recent introduction of NarrowBand Internet of Things (NB-IoT) technology in the 4th and 5th generations of mobile radio networks, the mobile communications context opens up significantly to the world of sensors. By means of NB-IoT, the mobile systems within 3GPP standardization introduce the peculiar functions of sensor networks, thus making it possible to satisfy very specific requirements with respect to those which characterize traditional mobile telecommunications. Among the functions of interest for sensor networks, the possibility of locating the positions of the sensors without an increase in costs and energy consumption of the sensor nodes is of utmost interest. The present work describes a procedure for locating the NB-IoT nodes based on the quality of radio signals received by the mobile terminals, which therefore does not require further hardware implementations on board the nodes. This procedure, based on the RF fingerprinting technique and on machine learning processing, has been tested experimentally and has achieved interesting performances.