Real-Time Indoor Localization and Tracking of Passive Targets by Means of Wireless Sensor Networks

Recently, the growing need of monitoring private or public areas for security purposes in civilian and military applications is driving the research community to design non‐invasive systems based on tiny sensing devices [1]. The tracking of a vehicle in a restricted area, the detection of animals in a dynamic environment, or the analysis of people behavior from movements are few examples of applications where the employment of systems for the localization and tracking of targets is mandatory. In the framework of wireless communications and technologies, the development of low‐power and low‐cost devices, such as Wireless Sensor Networks (WSN) [2], integrating on‐board processing and radio interface has favored the development of efficient cooperative signal processing algorithm for tracking purposes. Most of these systems are based on the processing of data acquired by dedicated sensor, or they assume to localize an active target, namely provided with some transmitting devices [3]. Unfortunately, in many applications the targets can not be equipped with wireless modules and the use of a complex system based on specific sensors is often not affordable. In this work, the localization problem is addressed by considering only the information provided by the quality indexes of the wireless links between the nodes of the WSN as in [4]. Consequently, unlike state‐of‐the‐art approaches, the infrastructure needed by the tracking procedure is limited to the nodes of the WSN, without the need of additional sensors. As a matter of fact, the target moving inside the scenario under test interacts with the electromagnetic signals transmitted by the wireless devices, thus modifying the values of the quality indexes measured at each node of the network. By reformulating such a problem in terms of a simplified electromagnetic inverse scattering problem, the localization and tracking of a passive target is carried out by means of a learning‐by‐example (LBE) strategy [5]. With respect to [4], the novelty of this paper lies in the application of the proposed approach to realistic indoor scenarios and in the use of differential measurements in order to remove the background contribution.