WiFi Identification of People Walking on the Other Side of a Wall is Now a Thing

Written by Bill Toulas
Last updated October 1, 2019

Researchers from the UC Santa Barbara have created a novel system based on a pair of WiFi transceivers, which can identify a person. To be more accurate, the system can correlate a person who appears in a sample video footage and then figure out if it is that same person who passes by at the other side of the wall. The researchers call this system “XModal-ID”, and they are confident it will open the way to a wide range of pioneering applications in security and surveillance.

The WiFi transceivers that are used in XModal-ID aren’t anything special, but regular off-the-shelf models. The system requires no previous training, as a video footage sample of the person that needs to be identified is enough. The 1488 tests that were carried out by the researchers featured an identification accuracy of up to 89% (average 84%), using video profiling samples for eight individuals. The main thing that identifies the people who pass by on the other side of a wall is the unique way they move their body parts when walking. The transceivers feed the recording of the signals to a computer which then analyzes them and figures out how a person moves.

The video feed that is provided as a sample is turned to 3D mesh, and then transformed into a WiFi signal with Born approximation. This creates a correlation platform that seems to be working quite reliably. Depending on whether the monitored area is outside or inside, the researchers used different methods to transform the signal into a usable format. For example, generating a WiFi spectrogram using Fourier Transform and Hermite Functions produced very reliable results during indoors testing. As one of the researchers stated: “Our approach is multi-disciplinary, drawing from areas of both wireless communications and vision.”

The implications of this new unit can potentially be staggering. Knowing that you are being monitored is one thing, but becoming the object of surveillance without realizing it is a completely different story. Someone could take your video footage and feed it to their XModal-ID system, and that would be enough to profile you forever. Then, they could set up WiFi transceivers that would constantly check for your presence on key location points, hidden behind walls, doors, ceilings, and even floors. The team presents an example of thieves being captured on video and profiled, and then recognized and arrested at a checkpoint, located somewhere else, in the future. As we recently discussed the Chinese super-camera, we’re entering an age where our right to privacy gradually becomes irrelevant, as we steadily lose the right to even know when it is being violated.

Are you worried about systems like the XModal-ID going mainstream, or are systems like this one making you feel safer? Let us know in the comments down below, or on our socials, on Facebook and Twitter.

For a better user experience we recommend using a more modern browser. We support the latest version of the following browsers: For a better user experience we recommend using the latest version of the following browsers: