WiFi can read through walls
news.ucsb.edu
news.ucsb.edu
Nakia wants 6G devices to act as 3D imaging clients, and for network operators to have access to a realtime, 3-dimensional visualization that can see right through walls.
this was literally used as a plot point in one of the batman movies a decade ago to highlight how dangerously invasive tech can be.
https://www.nokia.com/about-us/newsroom/articles/nokias-visi...
https://www.bell-labs.com/institute/blog/building-network-si...
> A very exciting innovation that 6G will bring to the table would be its ability to sense the environment. The ubiquitous network becomes a source of situational awareness, collating signals that are bouncing off objects and determining type and shape, relative location, velocity and perhaps even material properties. With adequate 6G solutions for privacy and trust, such a mode of sensing can help create a “mirror” or digital twin of the physical world in combination with other sensing modalities.
Here's a press release, for example: Nokia selected by Dedicado for 5G NSA network in Uruguay (https://www.nokia.com/about-us/news/releases/2022/12/08/noki...)
And a Linkedin profile (he must have a room in the office): Yuriy Pavlov - NSA Infrastructure Information Systems Manager. Nokia. (https://www.linkedin.com/in/yuriy-pavlov-1969904)
The popularity will explode the moment someone makes a commercial where someone is cooking with something messy and using their phone without ever touching it.
It's the people who make or break it. "This technology is too dangerous and must be destroyed, but hold on I need to use it first because it's really useful" doesn't feel like the strongest condemnation.
Why would I care that my neighbor knows we're home and moving around? If they really wanna creep on us, they can already use IR cameras or audio amplification or laser mics or whatever.
There's always a more sophisticated way to do something bad. Why would that be a good reason to lower the bar for doing it?
It's fine if I can take live 3D radar images of the inside of my bedroom. Not fine if someone else can.
When I was young my parents wanted me to learn about religions so I could make my own choice about which to follow. But I was constantly like "wtf how does this make sense???". And disruptive in class :) They had to take me out of Sunday school.
Another thing I never understood is predestination in calvinist religions, which are big in Holland. They believe it's already decided whether you go to heaven or hell, regardless of your actions during your life.
So how would you live your life? Personally I'd go for maximum enjoyment because whatever happens is already decided. But no, they're very strict and regimental in general. Never understood why. But of course I did only have Sunday school for a short while :)
> So how would you live your life? Personally I'd go for maximum enjoyment because whatever happens is already decided. But no, they're very strict and regimental in general. Never understood why. But of course I did only have Sunday school for a short while :)
I heard about the predestination thing well before I heard of Newcomb's paradox, but now I realise this is basically the same as "one box, or both boxes?"
That said it is definitely what any competent authoritarian would want.
I don’t want to control their discussion, it’s fine to discuss things and share different views. I want to understand why a group of people would welcome a panopticon because it would make people easier to control.
I don’t want to doxx myself so I’ll keep the details vague, but both of my parents came from disturbing environments and were subject to things that would universally be understood as deeply traumatic. I think they’re good people, but very very confused, and their fundamentalism was a product of the degree of unresolved trauma/dysfunction in their lives and their need to impose some kind of order on their circumstances.
Not all churches look this way. Mine sure did. Almost everyone had a story. Thankfully I escaped the bubble. But not without consequence. Through no fault of my own, I was indoctrinated into a belief system that I then had to spend decades and counting unwinding and unlearning. I mention this because a lot of people want to focus on other people’s lives because that’s what they were taught to do. People were born into beliefs they didn’t choose.
None of which is to excuse the behavior. But I’ve seen up close where it comes from.
Email me if you ever want to chat about this stuff.
Cell phone network operators need to know your location in order to route traffic to the correct tower in order to reach you, and that is not changing anytime soon. Same principle for WiFi beamforming.
Dont know if its still around.
But that's obviously false. They can know which tower to route your traffic to by the simple fact that that's where your traffic came from.
They know where you are because you broadcast simultaneously to all of their towers, but that's not something they need.
To get maximum data throughput, modern communications systems measure how the channel (ie. environment) changes the radio signals and then reverses those changes in the receiver. (search term: channel estimation). So it's inherent that a modern communications system is modelling its environment. A primary way communications systems become better is to increase the sophistication of their environment model.
Once the model becomes sophisticated enough, the only difference between a radar and a communication system is which information you choose to extract: the information the environment imprinted/modulated on the radio signal/carrier (ie. the model of the environment) or the information the transmitter imprinted/modulated on the radio signal/carrier (ie. the message). The first is radar, the second is communications. There is no technical reason why you can't extract both.
A pen can be both a writing implement and a deadly weapon. But until someone decides to use it as a deadly weapon, it's not one. Luckily we don't typically hear people explaining the anatomy of soft tissues in the human neck when talking about pens.
The problem is that this information is inherent to a communications system. It can't be separated. Communications is pervasive, so channel information is pervasive. How does one control something that is pervasive?
It's a Catch-22: Do we implement a surveillance apparatus to prevent people from accessing the channel information that can be used to build a surveillance apparatus?
thats the wrong question - simply remove all barriers that prevent consuners from inspecting inner working if their devices, and attach large statutory compensation in case a consumer privacy is violated.
In the movie Charlie’s Angels (2000) the supervillain technology allowed them to track the locations of the vast majority of people on the planet in realtime.
The real world has already jumped the shark for many different fictional supervillain doomsday technologies.
Which makes the whole discussion about privacy a joke because nobody can define what privacy is.
.... I think technology that enables visualization through walls is pretty obviously a privacy invasion. People are used to walls giving them privacy...
Simply by going about your day you publish metadata about yourself that can be used to derive behavior.
* https://ieeexplore.ieee.org/document/8904842
There is an IEEE Working Group to create an extension for 'active radar':
> IEEE 802.11bf will enable stations to inform other stations of their WLAN sensing capabilities and request and set up transmissions that allow for WLAN sensing measurements to be performed, among other features. WLAN sensing makes use of received WLAN signals to detect features of an intended target in a given environment. The technology can measure range, velocity, and angular information; detect motion, presence, or proximity; detect objects, people, and animals; and be used in rooms, houses, cars, and enterprise environments. The targeted frequency bands are between 1 GHz and 7.125 GHz (MAC/PHY service interface) and above 45 GHz (MAC/PHY).
* https://standards.ieee.org/beyond-standards/ieee-802-11bf-ai...
Presentation on the extension and Wi-Fi sensing:
5 GHz in particular has trouble passing through concrete and metal. If you have a small wood house or apartment it often isn't an issue, but if your office has a lot of concrete and steel, etc. it could be problematic. Those enterprise mesh networks exist for a reason :)
https://blog.ibwave.com/a-closer-look-at-attenuation-across-...
https://help.keenetic.com/hc/en-us/articles/213968869-Wi-Fi-...
5GHz was basically a no go. It wasn't a huge loss since few devices were running it back then.
So the most arduous part of the whole job appeared to be the demolition to start with. They started pounding into the thick plaster walls, and revealed a thick mesh of solid steel reinforcement. Of course, that took ages to take out, wherever they had to deal with such a wall.
And you know what? I was absolutely appalled by the chintzy material that went in to replace it. They just brought this drywall, you know, and I got to touch it and examine it, and I couldn't believe how flimsy it was, when compared to the reinforced plaster they'd just taken out. I mean, like, there were cost and time considerations, but it seemed like a travesty to me. Like we were sacrificing half of our solid, valuable home for a mere facsimile in the other parts.
Today, the WiFi is fairly bad. My dad only has the one router in a distant corner of the den. I don't know how reception is directly upstairs, where his bedroom is, but the downstairs bedroom can barely keep a signal going at the best of times.
That home renovation was happening around 1988, so possibly too early to really think of putting Ethernet in every room, but my dad would've been the dad to do it.
It's actually a fairly recent trend to not build load-bearing walls anymore and instead really on a skeleton. Walks can very well be constructed to be load-bearing. But, as you indicate, the resulting building would be more difficult to modify.
Exactly what are you getting by having the 'less flimsy' plaster and what are you losing with the 'flimsy' drywall? (I ask this as someone with three uncles that work in construction/renos.)
My walls are 1 inch of plaster with metal lath (vs the typical 1/2 thickness for drywall), on top of traditional wood slat lath, then insulation and studs. Typically if you have construction old enough for lath and plaster walls, they might also have independent studs per side, further isolating sound transfer.
It's also less porous and doesn't hold water like drywall, making it quite mold resistant.
2013 - gesture recognition from Wi-Fi https://wisee.cs.washington.edu/
2013 - see through walls with Wi-Fi https://people.csail.mit.edu/fadel/papers/wivi-paper.pdf
2023 - dense pose estimation from Wi-Fi https://arxiv.org/abs/2301.00250
It can't read 2-dimensional print on paper. Now that would be a massive (and terrifying) accomplishment
Interesting paper though, I wonder what non-infosec applications it might have.
Format is described on the page if you want to do it manually.
Of course the plaintext password is still encoded in the QR code data, so it's not very secure, but great for home guests.
Engineering ethics courses all boil down to "don't lie about stuff", not "think about the potential consequences of what you're building".
Usually, we can't rely on things simply not being invented, so we do the next best thing: invent it, and use the invented thing to learn how to manage the risk that thing introduced.
Here's another one: name the private (not gov contractor) orgs who are currently working on fission/fusion or even radiological weapons. There aren't any. For as disparate as the cultures of China and the US are, both realize the fact that DIY nukes are bad and universally disappear anyone going down that route.
UCSB paper: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=101... ("Analysis of Keller Cones for RF Imaging")
I’m not into RF so this is a bit imprecise, but it took like a RPi with GNU radio, a second router, and some code to diff the signals into a location.
The demos were compelling, you could see a heatmap of a user waving and so on. This is a product already sold.
So now we have to consider the wattage of a rotating system, and how fast does it need to rotate so the time not directly being radiated doesn't cause gaps in coverage. There's a lot of variable to cover here.
Wire up an RPI to a motor which reorients a 2.4ghz parabolic dish, first discovering the direction and elevation which best picks up the signal from the "scanner". Then, engage the old microwave emitter you've hooked up to the dish.
You will like as not fry their equipment. Bonus: the attacker may never have children.
Windows we used film, curtains shielded too. Two coats primer. Put copper stripts to ground as well on walls before paint. Didn't do floor so some rf gets in and out which is ok actually, you can create a focusing effect if you aren't careful about it.
If you want science, it's tough, nobody is going to fund it and it's harder and harder to find a control group.
There's studies of dubious quality if you Google emf and sleep.
Could you link to the film you used, and also the curtains?
Did this at all block your ability to use your own cellphone?
Somewhat related but I always thought a cool warfare device would be a grenade that contains an inflatable "soldier" with a built in heat signature. Toss them around for IR imaging spoofing.
Anyways..
https://link.springer.com/book/10.1007/978-981-99-2501-8
Any good research out there on WiFi, cellular and RF in general?
This is from the DIA but was acquired via a FOIA request so I don't know how valid it is. I would probably start looking at the references first.
The original Dune was released that year, and Terminator too...
Let me guess, this thing puts the wifi into some constant transmit mode and measures loss rates as it passes through objects, and scans in some pattern, measuring loss rates at the other side.
Is this just packet based radar? Is this a cute way of saying "using the 5GHz band to scan through walls when you have a receiver on the other side of your object"?
Here's the thing: different materials absorb / reflect radio waves with different characteristics. You could just as easily make these letters out of a completely different material that doesn't remotely resemble a letter and it could look exactly like a letter.
How is this useful? And why is wifi the medium to do it. You can't just do this with a regular old driver. Your normal driver will be selecting antennas, adjusting rates for loss rates. This is clearly a dedicated radio with some open source driver that they've hacked to do this.
Ugh. This just seems like a dumb waste of time. This isn't even a passive radar, as it requires there to be a receiver on the other side. There is no bouncing back of signals.
"It does not require any prior RF data for training a machine learning system for RF sensing." Yes, but it does require very specific material characteristics that it is trying to detect at fixed distances. Namely edge keller cones on specific materials, and hoping that other materials don't replicate similar patterns.
This sort of capability is being introduced into the newest WLAN standards and is being promoted as the next step towards smart homes and devices. The idea being that a standard router can be used to detect human presence, and therefore do things like turn on/off lights, HVAC, etc. This might make even more sense in the commercial space.
This would be as far as I'm aware the first attempt at doing something like this without additional hardware (mounted sensors). It should in theory lower the barrier to entry even further.
There are higher frequency variations of this that could in theory do things like detect breathing. Some people are talking about it being like the new lifeline button or baby monitor.
How well it's implemented will ultimately determine it's usefulness IMO.
From the way they described the goals of their research, it wasn't so much using WiFi as radar with specially flashed router drivers doing emitting and receiving, that's just a research convenience.
The whole point is to passively detect objects using already existing WiFi signals. It's easy enough to figure out where the router is, and if the router is regularly noisy enough, the idea is they would have enough data to work with.
The research results from a decade ago was able to pinpoint and uniquely identify humans and weapons. Not hard to guess how that could be useful to the DoD.