(Phys.org) —A team of researchers at MIT have been working this year on a system that can track people through walls with impressive accuracy using radio waves. The team showed the system earlier this month. IDG News Service made a video of the demo, which took place at MIT's Computer Science and Artificial Laboratory (CSAIL) in Cambridge, Massachusetts. The system is still in a proof of concept stage but the team spoke with reps from wireless and component companies during an open house recently. The system was developed by Professor and CSAIL Principal Investigator Dina Katabi and PhD student Fadel Adib. The technology uses low-power signals to track human movement and to decipher motions behind walls. Adib said their accuracy is higher than even state of the art Wi-Fi localization.
The approach involves three radio antennas—two transmitting and one receiving, pointed at a wall. In the demo, a person walked around the room on the other side of the wall. The system represented that person as a red dot on a computer screen. and could place the person on the other side of the wall n MIT project can track a user with an accuracy of +/- 10 centimeters. Earlier this year, another report from MIT that was following the CSAIL project's progress, noted that while researchers have long attempted to build a device capable of seeing people through walls, previous efforts involved expensive and bulky radar technology. The system at MIT uses low-cost technology. The goal is to come up with a device that is low-power, portable and simple enough for use to see through walls and closed doors.
Possible scenarios making use of such a system include law enforcement, to avoid personnel walking into an ambush; hostage standoffs; emergency responders trying to see through collapsed structures; and gaming. In addition, the system could be put to use for everyday needs in monitoring children and the elderly.
The IDG News Service video noted a few drawbacks in the system as-is. It can only track one person at a time; the rest of the area needs to be completely clear of movement. Another item on the to-do list is to make the system more compact.
Zach Kabelac, masters student, said in the IDG News Service video that "We can put a lot more work into miniaturizing the hardware. The antennas don't need to be as far apart as they are now. We can bring these closer together to the size of a Kinect, possibly smaller—in the process lose a little bit of accuracy but compared to the gains our system provides, it's minimal."
The group recently filed a patent, although there are no immediate plans for commercialization.
Explore further: Tailored 'activity coaching' by smartphone
More information: people.csail.mit.edu/fadel/wivi/design.html