The concept of X-ray vision – being able to see through walls, has been around for as long as we can remember – be it superheroes with X-ray vision or action-movie heroes with high-tech goggles. The revolutionary concept of X-ray vision was transformed into reality through the work of a group of researchers at MIT’s Computer Science and Artificial Intelligence Lab (CSAIL) led by Dina Katabi, an MIT Professor of Electrical Engineering and Computer Science and Fadel Adib, a Ph.D. student at CSAIL.
It all started in 2013 when the CSAIL team introduced WiVi, a device that used Wi-Fi signals to detect and track the direction of human movement through walls. Based on this research, they developed WiTrack later in the same year. Instead of Wi-Fi signals, WiTrack used very low-power radio waves that are 100 times smaller than Wi-Fi waves and 1000 times smaller than the ones that an average cell phone transmits. However, these radio waves enabled significantly higher accuracy and could track both two dimensional and three dimensional motions.
The team has raised the bar yet again with their recent unveiling of the RF capture at the SIGGRAPH Asia conference held from 02 - 05 November in Japan. The RF Capture uses radio waves that are 1000 times less powerful than Wi-Fi signals which allows the device to extract much more information with increased accuracy from them because of their structural design. The improved hardware and software designs make RF Capture a far more powerful tool than its predecessors overall.
"It basically lets you see through walls," said Fadel Adib, lead author of the 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."
The radio waves emitted by the device have very short wavelengths and thus can travel through walls. The waves are reflected back from objects blocking their trajectory. Since inanimate objects also reflect the signals, the system begins by identifying static objects and removing them from its analysis. Then, it takes a series of snapshots, looking for reflections that vary over time, which represent moving human body parts. The new system can identify up to 15 different people through a wall with 90-percent accuracy and track movements to within 0.8 inches. For moving people, the system is able to identify exactly which body part said person has moved with 99 percent accuracy.
The RF capture currently has a few limitations such as the fact that it can only track movements accurately when the subject is directly facing the device. The RF capture is also unable to perform full skeletal tracking as opposed to its motion detecting counterparts.
Emerald, an eldercare system that was developed using WiTrack can be enhanced with RF capture. Emerald was designed to predict, detect and track falls among elderly patients and also monitor heart and breathing rates remotely. This is scheduled to be the first commercial application of the RF capture. Additionally, the team expects that the device also has prospects in the sectors of video gaming and filmmaking.
Thumbnail Image: Super Friends Wiki, Wikimedia Commons
Cover Image: CSAIL, MIT