MIT Researchers have invented a new method, WiFi imaging, for utilizing WiFi signals to detect stationary objects hidden behind walls. WiFi signals have the ability to penetrate walls, and this breakthrough allows for the reconstruction of images of stationary objects positioned behind barriers using these signals.
While this development may spark concerns about privacy and potential misuse, it’s important to note that the technology is primarily designed for benign applications.
Detecting moving objects using WiFi signals has demonstrated promise in various applications. However, detecting stationary objects presents a distinct challenge as they lack the motion cues that facilitate tracking.
Researchers sought to overcome this challenge by using stationary objects, specifically English alphabets, as targets for their experiment. They developed an approach called “Wiffract,” which harnesses the radio waves emitted by off-the-shelf WiFi transceivers.
Wiffract leverages Joseph Keller’s Geometrical Theory of Diffraction (GTD), which exploits the unique signatures left by edges on the receiver grid. When a wave encounters an edge, it generates a cone of outgoing waves, known as a “Keller cone,” as described by GTD. This phenomenon is not limited to sharp edges but applies to all surfaces.
Using Coarse-to-Fine algorithm to capture these signatures, the researchers placed a receiver grid near the edge. The distinct patterns of rays left on the receiver grid were then used to construct the image of the object they were tracing.
Yasamin Mostofi, a professor at UC Santa Barbara, explained, “We then develop a mathematical framework that uses these conic footprints as signatures to infer the orientation of the edges, thus creating an edge map of the scene.”
This breakthrough enabled the researchers to achieve WiFi imaging of English alphabets concealed behind walls for the first time. They conducted experiments by positioning the letters of the word “BELIEVE” one by one behind a wall, and the results were impressive.
Not only were the letters readily identifiable, but the fine details of the letters were also accurately imaged. In essence, Wiffract has opened the door to WiFi’s ability to perceive objects through walls.
The research team conducted a series of 30 experiments to image uppercase English letters, consistently capturing the details of the letters with remarkable fidelity. Beyond reading letters, this technology holds promise for various applications, including crowd analytics, person identification, healthcare innovation, and smart space development.
Anurag Pallaprolu, the lead Ph.D. student on the project, emphasized that traditional imaging techniques struggle to produce high-quality images when using standard WiFi transceivers, particularly at lower frequencies, where surfaces may appear nearly specular and leave insufficient signatures on the receiver grid.
Wiffract’s innovative approach overcomes these limitations of this WiFi imaging, marking a significant advancement in WiFi-based imaging technology.