Researchers can see people through walls using routers and AI

With science, anything is possible! Researchers have managed to detect people through walls. Researchers at Carnegie Mellon University (Pittsburgh), specializing in political science and computer science, have developed this method to detect the three-dimensional shape and movements of the human body in a room.

According to Vice, they achieved this feat using routers and artificial intelligence. This allowed them to determine how many people were in a room and their exact position.

Thanks to the waves emitted by the network, the device made it possible to see through the walls since the bodies interfere with the waves. In this research, the bodies are very distinct from a table or a chair, since in fact, the Wi-Fi waves are disturbed by the bodies. Based on the DensePose system, developed by researchers from Meta, a neural network was developed to map the phase and amplitude of Wi-Fi signals sent and received by routers. This established a pattern capable of differentiating humans from a material thing.

For years, researchers have been working on techniques that can detect people without cameras or expensive LiDAR hardware, according to Capital. Already in 2013, researchers had used cell phone signals to see through walls, and in 2018 another from MIT (Massachusetts Institute of Technology) had already used Wi-Fi to detect men in another room and record their movements. This recently tested system would be more effective than a camera because it has no blind spots and could therefore detect people, even hidden behind a door or furniture. “It protects people’s privacy and it is a piece of equipment that can be purchased at a reasonable price,” the study authors explain. Indeed, most households have a Wi-Fi connection. This would therefore make it possible to “monitor the well-being of people or simply to identify suspicious behavior”, conclude the scientists.

With Capital