It may sound like a scene from a spy movie. Scientists at MIT have created a system that can penetrate through walls and track people’s movement.
The system developed by a team of computer scientists at MIT’s Computer Science and Artificial Intelligence Lab can beam out radio waves that bounce off human body, and receivers process the reflections using a computer algorithm to map movement in real time.
“It basically lets you see through walls,” said Fadel Adib, a Ph.D. student at MIT’s Computer Science and Artificial Intelligence Lab and lead author of a new paper describing the system. “Our revolution is still nowhere near what optical systems can give you, but over the last three years, we have moved from being able to detect someone behind a wall and sense coarse movement, to today, where you can see roughly what a person looks like and even get a person’s breathing and heart rate.” Livescience noted.
The system was unveiled at SIGGRAPH Asia conference and uses radio waves which are 1,000 times less powerful than Wi-Fi signals. Earlier the team led by Dina Katabi at MIT had developed a system that can recognize humans through walls using Wi-Fi signals. The radio waves are engineered specially to obtain more data at much weaker signal. The system uses a T-shaped antenna with four vertical transmitters and 16 horizontal receivers, which are controlled by a computer.
The way the system works is quite complicated. Since inanimate objects can also reflect radio waves, it starts by scanning the area for fixed objects and eliminate them for processing. Then it focuses on objects which vary over time by taking a series of snapshots, this represents moving body parts. But the moving body parts have to be at the right angle in relation to the antenna array for the sensors to detect the radio waves which are reflected; meaning each snapshot will only capture some body parts from frame to frame and the end result may not be recognizable. This is overcome by using an intelligent computer algorithm that has the ability to identify body parts from the snapshots and using a basic human structure, it can build a contour of the moving body. To further simplify the process of detecting a moving body in a wide space, the team uses a concept of military radar systems, that can lock and track objects, which improves the overall power consumption as well as image quality.
The team plans to use this technology in health care as an initial application and soon in motion and gesture control devices.
Country of Origin: United States of America