"Wiffract" Wi-Fi method can read letters through walls

Researchers have long explored the concept of using Wi-Fi technology to "see" through walls. A recent study proposes another approach to this dilemma, one that appears to be able to generate an approximation of the English alphabet using off-the-shelf, commercial Wi-Fi equipment.

Researchers at the University of California, Santa Barbara have developed a new method for imaging objects beyond the line of sight and named it "Wiffract." This technology is guided by the principles of Geometric Theory of Diffraction (GTD) and exploits the interaction between Wi-Fi radio frequency (RF) signals and the edges of the object to be imaged. With the right mathematical model, Wiffract can produce dramatic effects, such as "reading" shapes and letters through walls.

The researchers explained that according to the GTD principle, when a radio frequency wave encounters an edge point, an outgoing ray cone called a "Kellercone" will be generated. Wiffract's mathematical model can capture the edges of stationary objects using GTD theory and the corresponding Keller cones. Once "high-confidence edge points" are identified, Wiffract is able to reconstruct the object's shape while further enhancing the resulting edge map through advanced computer vision techniques.

According to the researchers, Wiffract has proven effective in a variety of experiments, including what they believe to be the first demonstration of Wi-Fi reading English letters through a partition wall. Key features of the new approach include its ability to utilize radio waves emitted by off-the-shelf Wi-Fi transceivers for imaging and the elimination of the need to train machine learning algorithms for radio frequency sensing.