Who needs a peep hole when a wifi network will do? Researchers from MIT have developed technology that uses wireless signals to see your silhouette through a wall—and it can even tell you apart from other people, too.
The team from MIT’s Computer Science and Artificial Intelligence Lab are no strangers to using wireless signals to see what’s happening on the other side of a wall. In 2013, they showed off software that could use variations in wifi signal to detect the presence of human motion from the other side of a wall. But in the last two years they’ve been busy developing the technique, and now they’ve unveiled the obvious — if slightly alarming — natural progression: they can use the wireless reflections bouncing off a human body to see the silhouette of a person standing behind a wall.
Not only that, the team’s technique, known is RF-Capture, is accurate enough to track the hand of a human and, with some repeated measurements, the system can even be trained to recognise different people based just on their wifi silhouette. The research, which is to be presented at SIGGRAPH Asia next month, was published this morning on the research group’s website.
So how does it work? It’s actually relatively straightforward: a device transmits wireless signals on one side of the wall, which propagate through it and are then reflected by bodies on the other side. The device then captures the reflections, which are passed to software to be cleaned up. As you might expect, this part requires some pretty serious processing, as different body parts, humans and objects introduce all kinds of interference.
Our IP Address: