Fine-grained human motion tracing — the ability to trace the trajectory of a moving human hand or leg, or even the whole body — is a general capability that is useful in a wide variety of applications. It can be used for gesture recognition and virtual touch-screens (e.g. Kinect-style natural user interfaces), activity recognition, monitoring of young infants and the elderly, or security applications such as intruder detection. Motivated by these applications, depth-sensing-based systems have been developed to implement motion tracing capabilities in cameras; however, these devices are limited because they have a constrained field of view (around 2 to 4 m range with a 60-degree aperture), and do not work in non-line-of-sight scenarios, preventing their use in many applications such as whole-home activity recognition, security, and elderly care.

The compact WiFi AP integrates WiDeo’s motion tracing functionality, and can reconstruct the hand movement made by humans in the living room. WiDeo traces motion even though the AP is separated by a wall and does not have a LOS path to the humans, and doesn’t require that the humans have any RF devices on them.

WiDeo is a device-free, compact motion tracing system that does not require the humans whose motion is being traced to have any devices on them, whether RFID tags or phones. WiDeo enables accurate, high-resolution, device-free human motion tracing in indoor environments using WiFi signals and compact WiFi radios. WiDeo mines the backscatter reflections from the environment that WiFi transmissions naturally produce to trace where reflecting objects are located and how they are moving.

Novel backscatter measurement techniques work in spite of the low band-width and dynamic range of WiFi radios. New algorithms separate out the moving backscatter from the clutter that static reflectors produce, and then trace the original motion that produced the backscatter, even if it has undergone multiple reflections. WiDeo accurately traces motion even when there are multiple independent human motions occurring concurrently (up to 5), with a median error in the traced path of less than 7 cm.

WiDeo, like a standard ToF camera, incorporates four main components:

  • Flash: This is the light source used to light up the scene; in WiDeo, this is simply the transmission that the AP in which WiDeo is housed is sending for standard communication. Wireless transmissions used for communicating packets act as the flash for the WiDeo.

  • Backscatter Sensor: This component looks at the backscatter arising from the environment when the AP’s transmission gets reflected and arrives back at the AP. The sensor teases out the individual signals emanating from each reflector in the environment as well as estimates each reflection’s intensity, angle of arrival, and relative time of arrival. The corresponding components in a standard camera are the image sensors that capture the light (aka the backscatter) from objects in the scene and form a picture of the scene.

  • Declutterer: The captured backscatter contains reflections from static objects that act as clutter to the reflections originating from the moving object WiDeo wishes to trace. The declutterer component figures out which of the reflections are from objects WiDeo doesn’t care about, and eliminates them so that motion tracing can focus only on reflections from moving objects.

  • Motion Tracer: This component looks at the reflections coming from moving objects over time to predict the actual physical motion that could have produced that sequence of reflections.

For more information, contact Evan Elder, Licensing Associate, at 650-725-9558.

Tech Briefs Magazine

This article first appeared in the December, 2017 issue of Tech Briefs Magazine.

Read more articles from this issue here.

Read more articles from the archives here.