Distance Based User Localization and Tracking with Mechanical Ultrasonic Beamforming

Many IoT smart systems such as smart homes require intelligent solutions to localize and track a user in a living space. Traditional solutions are either too simple in capability to reveal dynamic details or invasive in privacy leaking. This work designs and develops a novel noninvasive distance based user localization and tracking solution, DiLT, for smart systems. DiLT collects data by using commodity off-the-shelf ultrasonic sensors for minimal invasion on privacy, while adopting signal processing techniques to reveal detail dynamics embedded in the sensed data. DiLT consists of a mechanical ultrasonic beam-forming design for omni-space sensing, a contrastive divergence learning to localize a user and a binary back-off algorithm to track the motion of the user. It has been extensively evaluated in a research laboratory room and it demonstrates high effectiveness and accuracy in performance in the real environment.