WiFi signals identify walkers from their gait

Researchers in the US have used WiFi signals to detect the gait of people through walls and match it to video footage in order to identify individuals.  

According to those behind the technology, known as XModal-ID, it could potentially be used to identify criminals inside buildings where pre-existing video footage of the suspects is available. It works by measuring the way a person’s gait interferes with the signal from two WiFi transceivers. This signal is then compared to a simulated signal extracted and translated from pre-existing video footage.

Public WiFi could be used to screen bags for weapons and bombs

Crowd-counting WiFi technique looks through walls

“Our proposed approach makes it possible to determine if the person behind the wall is the same as the one in video footage, using only a pair of off-the-shelf WiFi transceivers outside,” said research lead Yasamin Mostofi, professor of electrical and computer engineering at the University of California, Santa Barbara.

“This approach utilises only received power measurements of a WiFi link. It does not need any prior WiFi or video training data of the person to be identified. It also does not need any knowledge of the operation area.”

From video footage, the system uses a human mesh recovery algorithm to extract the 3D mesh describing the outer surface of the human body as a function of time. Electromagnetic wave approximation then simulates the RF signal that would have been generated if this person was walking in a WiFi area.

Next, a time-frequency processing approach extracts key gait features from both the real WiFi signal and the simulated signal from the video. The two signals are then compared to determine if they match. In experiments, the technology successfully identified subjects between 80 and 90 per cent of the time. The system builds on previous work in the Mostofi Lab, which has investigated sensing with radio frequency signals since 2009.

“However, identifying a person through walls, from candidate video footage, is a considerably challenging problem,” said Mostofi. “The way each one of us moves is unique. But how do we properly capture and compare the gait information content of the video and WiFi signals to establish if they belong to the same person?”

The proposed methodology and experimental results will appear at the 25th International Conference on Mobile Computing and Networking (MobiCom) on October 22.