Object Tracking Through RSSI Measurements in Wireless Sensor Networks

In this paper, the localization of moving and transceiver-free objects is addressed by processing the Received Signal Strength Indicator (RSSI) available at the nodes of a Wireless Sensor Network (WSN). Starting from the RSSI measurements, the probability of the presence of unknown mobile objects is determined by means of a customized classification approach based on a Support Vector Machine (SVM). Experimental results assess the feasibility of the proposed approach. The need to deal with real-time localization in innovative civilian and military applications has caused a growing interest in suitable signal processing techniques and wireless communication systems.