Impact of body wearable sensor positions on UWB ranging

Abstract

In recent years, Ultrawideband (UWB) has become a very popular technology for time of flight (TOF) based localization and tracking applications but its human body interactions have not been studied yet extensively. Most UWB systems already proposed for pedestrian ranging have only been individually evaluated for a particular wearable sensor position. It is observed that wearable sensors mounted on or close to the human body can raise line-of-sight (LOS), quasi-line-of-sight (QLOS), and non-line-of-sight (NLOS) scenarios leading to significant ranging errors depending on the relative heading angle (RHA) between the pedestrian, wearable sensor, and anchors. In this paper, it is presented that not only does the ranging error depend on the RHA, but on the position of the wearable sensors on the pedestrian. Seven wearable sensor locations namely, fore-head, hand, chest, wrist, arm, thigh and ankle are evaluated and a fair comparison is made through extensive measurements and experiments in a multipath environment. Using the direction in which the pedestrian is facing, the RHA between the pedestrian, wearable sensor, and anchors is computed. For each wearable sensor location, an UWB ranging error model with respect to the human body shadowing effect is proposed. A final conclusion is drawn that among the aforementioned wearable locations, the fore-head provides the best range estimate because it is able to set low mean range errors of about 20 cm in multipath conditions. The fore-head's performance is followed by the hand, wrist, ankle, arm, thigh, and chest in that order.