Experimental investigation of wireless sensor networks for fingerprinting-based posture recognition

In this paper, we consider a radio fingerprintingbased localization system for indoor posture recognition applications. Fingerprinting allows target localization on the basis of radio-frequency measurements of the Received Signal Strength
(RSS), taking into account the presence of fading by means of a training phase. Posture recognition is then performed by localizing, through fingerprinting, a group of targets placed on the portion of interest of the user body—in this work, we will focus on arm posture. Our results show that the proposed system guarantees an accuracy of a few centimeters. Experimental results, obtained with a SunSPOT wireless sensor network, are confirmed by analytical results.