As you know, a Wi-Fi signal can pass through any (well, almost any) obstacles, and German scientists were able to use this property to learn to “see” through the walls.
The idea is based on the analysis and registration of forms created by electromagnetic waves that “bounce” from objects while moving around them. That is, in simple terms, scientists have learned to recognize objects that “sweep” waves of Wi-Fi signal even through walls.
How it works?
The study, related to the method of 3D-visualization of these signals, began as a student thesis project, then to go into a larger study – this follows from a publication in the Physical Review of Letters. The technique described in the study is capable of providing visualization at a frequency of 10 times per second and recreating the internal structure of the whole building in large-scale simulation.
Using Wi-Fi to work with visualization is not a new concept, but the authors of the document argue that for the first time signals are used to create 3D holograms of large spaces. The system does not yet have enough accuracy to reflect many details, but it is enough to distinguish between the main objects in space.
“If there is a cup of coffee on the table, you can see that there is a small object on the table, but you will not be able to recognize its distinct form,” said Philippe Hall, a student of physics at the Technical University of Munich, who co-authored the study. “But you can fully recognize a person or, for example, a dog, but in fact, any object measuring more than four centimeters.”
Hardware and Algorithm
The method uses a Wi-Fi transmitter as a low-power radar system, with two antennas: one of them forms a 2D plane, and the other fixes a signal with respect to the first one.
Once the antennas collect enough data to form an image, the three-dimensional representations of the objects are transferred to a digital reconstruction algorithm that creates a holographic map of objects in space. According to Hall, the scanning system will eventually become even faster and more accurate. How the system works can be seen in this video:
