Sign In

Communications of the ACM

ACM TechNews

Footstep Sensors Identify People by Gait

View as: Print Mobile App Share: Send by email Share on reddit Share on StumbleUpon Share on Hacker News Share on Tweeter Share on Facebook
A woman walking.

A team of researchers has developed remote sensors that analyze footsteps by measuring minute floor vibrations.

Credit: Thinkstok

Researchers at Carnegie Mellon University (CMU) have developed sensors that analyze footsteps by measuring minute floor vibrations.

The vibrations can be used to identify specific individuals, and to test a new method of hands-off health monitoring.

The sensors are cylindrical devices a few centimeters tall that sit on the floor and can sense a walker up to 20 meters away.

The sensors are distributed as an array throughout the area where they are to detect footsteps.

The sensitive detectors pick up a lot more noise in a busy building, so the researchers had to teach the system to distinguish footsteps from background noise.

Said researcher Hae Young Noh, "We do various signal-processing and machine-learning [techniques] to learn what is the human-related signal versus other noise that we’re not interested in."

From Scientific American
View Full Article


Abstracts Copyright © 2020 SmithBucklin, Washington, DC, USA


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account