Massachusetts Institute of Technology (MIT) researchers published a paper earlier this week that demonstrated how specific radio signals can be used to track human beings through walls.
The paper, “Capturing the Human Figure Through a Wall,” describes the five researchers' system, RF-Capture, which they say operates similarly to radar and sonar. The group's work differs from prior research in that the subject doesn't have to wear a sensor to be detected. Instead, the tool relies on a sensor a room away, separated from the subject by a wall.
RF-Capture has about a 96 percent accuracy rate when distinguishing between five different people, and a rate of approximate 88 percent for 15 users. The system can also identify which body parts a person is moving with 99 percent accuracy when the user is three meters, or nine feet, away from the sensor. It has 76 percent accuracy when the person is eight meters, or 26 feet, away.
To earn its high accuracy rate, the system transmits wireless signals through the wall, which then reflect off a person's body and back to the device. A press release on the new technology notes that the researchers' primary challenge was how to differentiate individuals and their body parts. That's where algorithms come into play.
The sensor first scans the room to capture wireless reflections off objects, the press release states. The tool simultaneously monitors how the reflections vary as a person moves throughout the room. This data is then used to create a human body silhouette and unique “fingerprints” for each person.
The technology could be used for emergency situations, including if someone falls down in a home and is unconscious. It could also be incorporate into gaming, the researchers said.
“In the same way that cellphones and Wi-Fi routers have become indispensable parts of today's households, wireless technologies like this will help power the homes of the future,” Dina Katabi, MIT professor and paper co-author, said in the university's press release.
