Humans Can Be Tracked With Unique 'Fingerprint' Based On How Their Bodies Block Wi-Fi Signals (theregister.com) 22
Researchers from La Sapienza University in Rome have developed "WhoFi," a system that uses the way a person's body distorts Wi-Fi signals to re-identify them across different locations -- even if they're not carrying a phone. By training a deep neural network on these subtle signal distortions, the researchers claim WhoFi is able to achieve up to 95.5% accuracy. The Register reports: "The core insight is that as a Wi-Fi signal propagates through an environment, its waveform is altered by the presence and physical characteristics of objects and people along its path," the authors state in their paper. "These alterations, captured in the form of Channel State Information (CSI), contain rich biometric information." CSI in the context of Wi-Fi devices refers to information about the amplitude and phase of electromagnetic transmissions. These measurements, the researchers say, interact with the human body in a way that results in person-specific distortions. When processed by a deep neural network, the result is a unique data signature.
Researchers proposed a similar technique, dubbed EyeFi, in 2020, and asserted it was accurate about 75 percent of the time. The Rome-based researchers who proposed WhoFi claim their technique makes accurate matches on the public NTU-Fi dataset up to 95.5 percent of the time when the deep neural network uses the transformer encoding architecture. "The encouraging results achieved confirm the viability of Wi-Fi signals as a robust and privacy-preserving biometric modality, and position this study as a meaningful step forward in the development of signal-based Re-ID systems," the authors say.
Researchers proposed a similar technique, dubbed EyeFi, in 2020, and asserted it was accurate about 75 percent of the time. The Rome-based researchers who proposed WhoFi claim their technique makes accurate matches on the public NTU-Fi dataset up to 95.5 percent of the time when the deep neural network uses the transformer encoding architecture. "The encouraging results achieved confirm the viability of Wi-Fi signals as a robust and privacy-preserving biometric modality, and position this study as a meaningful step forward in the development of signal-based Re-ID systems," the authors say.
You chose to invent this RIGHT NOW? (Score:3)
https://mastodon.social/@alexa... [mastodon.social]
Does RFID blocking clothing/accessories help? (Score:2)
And where's the medical uses of this kind of tech??
Re: Does RFID blocking clothing/accessories help? (Score:3)
You want something that changes your signal over time, or at least, changes it since the last time you were scanned.
Look for my soon-to-be-released anti-WiFi-fingerprinting underwear. Comes in a pack of 50, each with different positioning of reflecting and absorbing materials. Each pair makes you look different to the scanner.
Sorry, privacy-preserving? (Score:5, Insightful)
> confirm the viability of Wi-Fi signals as a robust and privacy-preserving biometric modality
How exactly does this preserve privacy?
Re: Sorry, privacy-preserving? (Score:2)
Re: (Score:3)
We all know that doesn't work.
A person can be tracked using this unique identifier until they buy something (with non-cash payment that is traceable), and then they're identified. They can then be tracked across other locations, and identified there.
It's the same problem as face recognition and other unique identifiers being used to track people.
My Wiz Wifi bulbs utilize this technology (Score:2)
To provide motion sensing. It requires a group of bulbs to be connected to the same Wifi AP, though.
Meanwhile, when I do a backup of my HTPC upstairs to my NAS downstairs over Wifi, the throughput changes from 350 to 500 Mbps when I move. But not the way I expect : it goes to 500 Mbps when I'm sitting and in the way. And 350 Mbps when I go to the bathroom.
Definitely easy to pick up signals.
Re: (Score:2)
So how does sensing work? I'm assuming that the Wi-Fi signals need to pass between an access point and some type of sensor. This means that the sensing only works at a limited number of locations because there are a limited number of sensors, but this is still okay if the number of entrances and exits is small.
More importantly, I wonder if the positioning and orientation of a person between the Wi-Fi antenna and the sensor matter. If the person walks sideways or upside down or jumps off the ground, are the
I'm getting some piercings (Score:2)
I think I'll get some hoop earings that resonate at 2.4 and 5Ghz. Maybe with a diode and capacitor in them so they act as very noisy reflectors.
Re: (Score:2)
The great thing about tinfoil hats is that you can remodel them to change your WiFi signature very easily.
Joking aside, it's a shame most people stopped masking after the pandemic. As well as preventing the spread of respiratory diseases, it was good for privacy in an age where face recognition is so cheap.
~1 in 20 does not a "fingerprint" make (Score:1)
A person's actual fingerprints are either unique or very close to unique.
This is about 1 in 20, which is far from unique.
As others have pointed out, it's also easy to change.
Unique? (Score:2)
95% is only 19 in 20. That's far from demonstrating "unique". Also, anything metal on your person - change, keys, etc. - will effect your "signature". Maybe you could follow a person around the office on a particular day, but that person will probably appear different the next day.
Re: (Score:2)
Wait till you hear about the dataset it applies to:
"The dataset collects the CSI measurements of 14 different subjects. For each subject, 60 samples were collected while they were performing a short walk inside the designated test area. The samples were collected in three different scenarios: subjects wearing only a T-shirt, a T-shirt and a coat, and a T-shirt, coat, and backpack, respectively"
95% accuracy on that is interesting, but definitely not 95% generalized on all humans.
Re: Unique? (Score:2)
Yes we've all watched Continuum (Score:2)
Right?
I missed something there (Score:2)
How exactly did they got from a thing that can identify and pinpoint you just by your body, without using cameras, using a piece of equipment which is ubiquitous in all civilized parts of the World to a thing that is a "privacy-preserving biometric modality"???
Or is it just a 2025 dystopian version of the old /. meme:
1. Do stupid thing
2. ?????
3. profit?
95.5% (Score:2)
Is that better than police accuracy?
Just keep keys in your pockets... (Score:2)