ZLT say:

With the help of Wi-Fi and artificial intelligence, you can see through walls and spy on anyone

overclockers.ru

3 min

View Original

Researchers at the University of California, Santa Barbara have made a significant breakthrough: Wi-Fi signals can be read through walls, representing major implications for the future of wireless communications and security.

Researchers working at , have unveiled a revolutionary method that allows Wi-Fi signals to penetrate walls while tracking the intricate details of objects behind them. , not only demonstrates the untapped potential of wireless signals, but also opens up a variety of possibilities, from surveillance to medical diagnostics.

The research focuses on the interaction of edges with incident waves using geometric diffraction theory (GTD). The proposed method, called "Wiffract", maps the edges of stationary objects using GTD and corresponding Keller cones.

Building on the foundational work of the Mostofi laboratory, which has been at the forefront of sensor technologies using everyday RF signals such as Wi-Fi since 2009, Wiffract expands the horizons of application. The technology's capabilities range from crowd analysis and people identification to advances in smart healthcare.

Detailing the technical intricacies, Mostofi: “When a wave hits an edge, it creates what we call a 'Keller cone' based on the geometric theory of Keller diffraction (GTD). Edge orientation affects the resulting cone footprint on the receiver reticle. Using these traces, we created a mathematical model that deciphers these signatures, allowing us to determine edge orientation and subsequently create a detailed edge map of the entire scene.”

The general idea stems from the fact that Wi-Fi signals essentially bounce around and off objects. Using a receiver grid allows movement to be tracked along the edges of an object.

This approach allowed Wi-Fi to read the English alphabet through walls, revealing the intricate details of the letters. However, the team indicated that the technology could be used to create maps of indoor places, track and monitor human body movements, monitor public transport, and a range of other applications.

“Depending on the orientation of the edge, the cone leaves different marks (i.e., conical sections) on the receiver reticle. We then develop a mathematical framework that uses these conical traces as signatures to determine the orientation of the edges, thus creating a map of the edges of the scene,” Mostofi explained.

This isn't the first time Wi-Fi has been used to "see" through walls. Last year, this drone used a Wi-Fi connection protocol to locate devices such as mobile phones, smartwatches, laptops and tablets in a room, and then created a map of the location. The drone uses a connectivity feature known as “polite Wi-Fi,” which allows even password-protected devices to respond to contact attempts if they are within range. By sending multiple messages to a device and measuring the response time, the drone can pinpoint the location of each device to within a meter. Dr. Ali Abedi, a professor of computer science at Waterloo, highlighted the potential security applications and risks of such technology, noting that it could be used to track movements inside buildings or identify smart devices in the home.

Although Wi-Peep uses a completely different process and technology to "see" through walls, the overall security and privacy implications are enormous.

Law enforcement could potentially use the technology to track criminal activity without physically intervening, and hospitals could monitor patients without wearable medical devices, but it also raises concerns about warrantless surveillance and privacy. Technology's ability to identify people by their silhouette or even the way they walk, as evidenced by its use in personal identification, is both impressive and troubling.

In the wrong hands, it can be used to track people without their knowledge, which could lead to potential stalking or identity theft. Moreover, if integrated with existing data collection methods, it can provide a more complete profile of an individual, leading to increased data privacy concerns.