Take a closer look: Researchers have been exploring the concept of using Wi-Fi to “see” through walls for some time. A recent study now suggests an alternative method to address this challenge, which appears to be capable of producing an approximation of the English alphabet using readily available commercial Wi-Fi devices.
Researchers at the University of California, Santa Barbara have developed a novel method for imaging objects beyond the line of sight, which they call “wiffract.” This technique utilizes the interaction of Wi-Fi radio frequency (RF) signals with the edges of objects that need to be imaged, guided by the principles of geometric diffraction theory (GTD). With the appropriate mathematical model, Wiffract can achieve remarkable results, such as “reading” shapes and letters through walls.
Researchers explain that when an RF wave hits an edge point, it creates a cone of outgoing rays known as a “Keller cone” according to GTD. Wiffract’s mathematical model can capture the edges of stationary objects by leveraging GTD theory and the corresponding Keller cones. Once Wiffract has identified “high-confidence edge points,” it can reconstruct the shapes of objects while further improving the resulting edge map through advanced computer vision techniques.
According to the researchers, Wiffract has proven effective in various experiments, including what they say is the first demonstration of Wi-Fi reading the English alphabet through walls. Key features of this new method include the ability to use radio waves from commercially available Wi-Fi transceivers for imaging and the fact that it no longer requires training a machine learning algorithm for RF detection.
Thanks to the ubiquity of Wi-Fi and other wireless signals, the team explains, there is now “significant interest” in using radio signals for various applications, including sensing and “learning about the environment.” Previous imaging methods relied on motion for “activity detection” or person identification, while imaging the details of stationary objects remained a significantly challenging problem.
Wiffract offers a solution to this problem as it can effectively map stationary objects over Wi-Fi, which could be valuable for “scene understanding and contextual reasoning in general.” The researchers suggest several potential applications for this new technology, including smart homes, “smart spaces,” structural health monitoring, search and rescue operations, surveillance, excavation sites, and more.
The UC Santa Barbara team avoids any discussion of Wiffract’s privacy implications. Technology that can read through walls could have serious consequences Safety concernsThis potentially offers cybercriminals a new tool to remotely compromise home privacy. Law enforcement agencies may also use this technique, hopefully for legitimate purposes.