Scientists Develop Wi-Fi Method to Identify People Through Walls

A research team from the University of California, Santa Barbara, led by Professor Yasamin Mostofi, has successfully developed a method to identify a person located behind a wall using Wi-Fi signals and video recordings for the first time.

The new technique, called XModal-ID, involves using a Wi-Fi transmitter and receiver along with a previously recorded video of a person’s movements. The video is used to create a three-dimensional model of the person, which helps simulate how the Wi-Fi signal reflects off their body and to generate a reference chart of these signal changes.

This reference chart allows researchers to identify a specific person behind an obstacle. The Wi-Fi transmitter sends a signal through the wall, which is then captured by the receiver after it bounces off the person. The changes in the reflected signal are compared to the reference chart to determine the person’s identity.

The XModal-ID method was tested on eight participants. The accuracy varied depending on the material of the obstacle, with an average accuracy of 84%.

Researchers believe this new method could help law enforcement, for example, by identifying a suspect who has previously appeared on surveillance cameras inside a building.