We're getting kick-ass at seeing through walls using just Wi-Fi – MIT Eggheads at MIT are getting better and better at using Wi-Fi signals to see through walls, capturing motion accurate to within a few centimetres. Using the same signals sent out by your wireless router, the team built a device it has called RF-Capture, and placed it behind an ordinary lab building wall. It was able to pick up …
10 years ago, I'd have been excited about something like this. Now, my first thought is "wow!" but only for a fraction of a second. Then I start thinking about how Govt. and security services can use this against us for even more intrusive monitoring and surveillance :-(
Even with lath and plaster walls like I have I still get wifi out and LTE in, because I have windows in my house. Most people cover their windows with blinds or curtains, not metal, so you'd be able to use this or similar technology to spy on people at least through their covered windows even if their walls blocked the signals.
Come to think of it my outer doors are solid wood with no metal aside from the hinges, so they'd be mostly transparent to RF as well.
>None of the images seems capable of tracking their legs, and they all seem to be wearing jeans.
Legs are curved. The areas of the body this system 'sees' most clearly are flat and head-on to the sensor.
Did you notice that the people in the video all held their hands open, with fingers together and facing the camera? That clinched it for me.
If this system fired ping-pong balls at the subject and counted the ones that came directly back to it, the resulting heatmap wouldn't look too different to this team's images.
Good work!
This post has been deleted by its author
>They've got flat arms and heads?
Yes, in relation to the frequencies of RF used. From the MIT paper:
The challenge in using RF to capture a human figure is that not all body parts reflect the signal back to the sensors. Specifically, at frequency ranges that traverse walls, human limb curves act as ideal reflectors; hence, they may deflect the signal away from the sensors rather than back to them. (This is because RF signals that traverse walls have a wavelength of multiple centimeters, which is larger than the surface roughness of human body parts, causing each part to act as a perfect reflector [Beckmann and Spizzichino 1987].)
So, it is the geometry of the body, not the presence of blood vessels etc that this system works on.
I'm off to file a funding application for a big machine that fires ping pong balls at engineering students. For a smaller grant I'd be willing to use sociology students instead.
Actually... I may just go buy 10,000 ping pong balls and make the machine out of Lego for my own entertainment...
>They've got flat arms and heads?
There are flatter, harder parts of the head that are facing the sensor, yes.
The hands show up when they are held facing the camera.
>The image 'hot spots' around blood rich areas like the head and chest,
The head and chest are also thicker than the limbs. Shoes show up well.
>If this system fired ping-pong balls at the subject they'd bounce off the wall. That's the point of the research, it "sees" through walls.
Okay, you took the analogy too far. I was trying to convey the basis around which 'stealth' vehicles are designed... it's about the shape. If you drop a ball onto a flat floor, it will come back to your hand. If you drop a ball onto a curved surface, it will likely bounce away. Legs are Cylindrical, chests and heads present some flat area facing the sensor. Flat, thin hands show up better than thicker rounded arms.
Nobody else seems to have considered the main application - recording body location, movement and posture through a bedroom wall. Yeah, probably just me.
Should be good, presumably, anywhere a WiFi signal can be received through a wall. Including apartments and semi-detached and terraced houses. I may just have to flatten out my tinfoil hat and use it to paper the outside walls.
Which leads me on to wonder what impact the foil covering on insulation such as Celotex has on radio waves, and if this isn't much then do I need an alternative to tinfoil for my hat? How do you signal proof party walls, and ceilings and floors in multi-storey dwellings?
RF-proof wallpaper might becone a big seller. Now, why has my mobile phone stopped working?
Ummm.....if this is based on reflections then can you use mobile phone transmissions to map through walls? Why am I calling you? No real reason, just checking up on the strange sounds from next door....
Time for the medication, probably.
There's an insulation product sort of like bubble wrap but made of foil that I stuck up in a loft once. Difficult to know if it had an effect on the mobile signal in the house , as the signal in the area was shit anyway.
Although presumably you could get some shredded foil, but like those separate strands of tinsel that get bloody everywhere, stick it to a sheet of lining paper, then that to a wall, then ordinary wallpaper on top. Might just have enough of an effect to render this sort of system ineffective. Bit like chaff really.
But what happens if you have half metre thick solid stone walls and the windows have a sprayed on coating that is designed to heat from the sun? In my house I have one room that has ordinary glass in the window and I can get a mediocre phone signal in there or I have to go out on the terrace for a good phone signal. My wifi is not available outside the building.
> Seems like a great tool just imagine now crooks can make sure no one is home before they break into houses.
In my housebreaking days, that great tool already existed and was called a door bell (occasionally, fixed line phone rang from the booth across the road).
Why are amateurs always trying to make things complicated?
> This will allow you to determine exactly where to release the knockout gas so you can plunder with impunity.
That exposes you to much more serious charges than a simple break-in (especially one where you do not have to make an entry by force, e.g., through an open door or window) thus making the project non cost effective.
As I said, only amateurs think like that. Please leave the housebreaking to us professionals--people like you are giving us a bad name.
we all have to wear stealth clothing. It comes in black only and is specially shaped to avoid reflecting radio waves.
Might be in high demand for motorbike riders, as well. Imagine the frustration of the cops waving their radar gun and not getting any speed information.
A stealth motorbike with no flat surfaces and a RF absorbing coating would be required as well (you do know that historically most of the people who worked with those extremely expensive exotic materials are in poor health right now). Maybe the latest generation of RF absorbing materials are not as toxic, but they are still probably top secret.
Be able to work out where the criminals are when breeching a house, or dealing with a hostage situation.
Unless the crims are smart, and turn off their wifi/mobile phones
Could be good for a police sniper in the US though - high powered rifle capable of easily piercing wooden walls, use this to see through the walls, take the shot - no longer need an open window or clear line of sight.
Would be interesting to see how different types of walls affect the outcome (here in the UK, its mostly double brick, often with cavity insulation)
To be honest, any attempt to hide an individual is probably going to make things worse. The whole point to the technique is that the technology detects that the signal is reflected differently to just an empty room, and is fast enough to track differences in the form of movement. As such metal suits, chain mail, tinfoil hats or even denim underwear would just alter what is being monitored.
Probably the only real protection would be stopping the signal from getting through altogether, such as converting your dwelling to a Faraday cage. Might make phone calls tricky, though.
Although I'm at work and therefore watched that with the sound off, judging by the animated arrows alone it looks like the system both transmits and receives dedicated wifi signals (i.e. it's active radar), rather than uses an existing wifi box somewhere else (passive radar).
Can someone set me straight, preferably in the most condescending way possible?
I thought it was meant to receive and render only, using an array/matrix of receiver/sensor/antennas, if you will. And not a system with both the source signal and the receiver matrix as one unit. Consider the hotspot as the source of the signal, much like a very bright lamp with no soft diffusion glass, just a bright signal. So you place a receiver matrix on the side of the building and collect the signals en mass of the transmitting router and then pick through the signals that are reflected or otherwise altered by objects blocking the source. My only question would be; what if you moved the router to the wall next to the "sensor wall"? Will the antenna matrix still be able to pick out the objects via reflection when the source is in-between, or does it blind it? And what about Rimmer? How does Hard Light Hologram tech work? Is the Light Bee just a display device and requires a connection to Holly/Hilly? And what about an Apple Watch face of Hilly? Why can't I get one of those? This future sucks.
"Through the wall" active ultra-wide-band microwave sensing has been permitted only to police and public safety agencies on privacy grounds. Considering that much of what UWB does can be accomplished with (basically) a few extra antennas and some software, this could could break even that slight protection.
O, rave new whirled, that had such critters in it.
- Spl Czhkcr
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
