Slashdot Mirror


MIT Researchers Can See Through Walls Using Wi-Fi

itwbennett writes "MIT Professor Dina Katabi and graduate student Fadel Adib have developed a system they call Wi-Vi that uses Wi-Fi signals to visualize moving forms behind walls. How it works: 'Wi-Vi transmits two Wi-Fi signals, one of which is the inverse of the other. When one signal hits a stationary object, the other cancels it out. But because of the way the signals are encoded, they don't cancel each other out for moving objects. That makes the reflections from a moving person visible despite the wall between that person and the Wi-Vi device. Wi-Vi can translate those faint reflections into a real-time display of the person's movements.'"

75 comments

  1. its called radar by Anonymous Coward · · Score: 5, Informative

    It's not exactly new either. The only difference here seems to be that the radar signal source is just a low power wifi AP. Yawn.

    1. Re:its called radar by K.+S.+Kyosuke · · Score: 2

      MIT has already been doing some nice experiments with WiFi/microwave oven/ISM band radars. I wonder if this isn't an extension of their previous projects. Using unmodified WiFi electronics seems to be a neat trick, though.

      --
      Ezekiel 23:20
    2. Re:its called radar by Anonymous Coward · · Score: 5, Informative

      Standard radar cannot see through walls because the receiver is overwhelmed by the wall reflection (its ADC is saturated). According to the authors, this is a very well known challenge and is called the "Flash" effect. Wi-Vi's new nulling algorithm solves that problem, enabling for the first time narrowband RF to overcome this effect. This link explains their technique: http://people.csail.mit.edu/fadel/wivi/design.html

    3. Re:its called radar by K.+S.+Kyosuke · · Score: 1

      A Through-Wall Gesture Interface

      Wi-Vi leverages its ability to track motion to enable a through-wall gesture-based communication channel.

      That will be great in hostage situations for the cops to check whether the perps are giving them the finger.

      --
      Ezekiel 23:20
    4. Re:its called radar by peragrin · · Score: 2

      they can tell the difference between moving and non moving objects.

      The next version will be able to tell the difference in general materials, brick, stone, sheetrock, wood, warm bags of no longer potable water.

      The version after that will introduce portability. and thus the tricorder will be born.

      --
      i thought once I was found, but it was only a dream.
    5. Re:its called radar by Anonymous Coward · · Score: 0

      That will be great in hostage situations for the cops to check whether the perps are giving them the finger.

      ...and for the kidnappers to know where the cops are assembling.

    6. Re:its called radar by Anonymous Coward · · Score: 0

      Pfft... I can see through walls using only a hand drill.

    7. Re:its called radar by Anonymous Coward · · Score: 0

      err......Range Bin?

    8. Re:its called radar by Anonymous Coward · · Score: 0

      Holy crap you're right, this whole time that never occurred to me!
      -MIT Person Involved In Doing This

  2. Misleading headline by Anonymous Coward · · Score: 2, Insightful

    Headline should have been "MIT Researchers Can Track Movement Behind Walls Using Wi-Fi".

    1. Re:Misleading headline by gl4ss · · Score: 2

      Headline should have been "MIT Researchers Can Track Movement Behind Walls Using Wi-Fi".

      but then it sounds suspiciously similar to an older headline on the same subject. and is rather boring.

      --
      world was created 5 seconds before this post as it is.
    2. Re:Misleading headline by Anonymous Coward · · Score: 0

      With enough routers in a grid pattern, you may actually be able to create a picture - based on the difference in signal reflection. Say... a truck load of microcells? The NSA, CIA, and OtherAs would probably love it.

    3. Re:Misleading headline by squiggleslash · · Score: 1

      I'm not even sure that's accurate. One signal is the "inverse" of the other? Then how is it Wi-Fi?

      --
      You are not alone. This is not normal. None of this is normal.
  3. Code source or it didn't happen by kk49 · · Score: 2

    Gimme! Please :)

    --
    You can have your god back when you are old enough to handle the responsibility.
    1. Re:Code source or it didn't happen by kk49 · · Score: 5, Informative
      --
      You can have your god back when you are old enough to handle the responsibility.
    2. Re:Code source or it didn't happen by kk49 · · Score: 2

      It was done with USRPs (http://www.ettus.com/ which are awesome) and one of the WiFi encoding schemes, not with standard WiFi hardware :(

      --
      You can have your god back when you are old enough to handle the responsibility.
    3. Re:Code source or it didn't happen by citizenr · · Score: 2

      you will need 3 x USRP first

      Its not WiFi. They could of named it seeing through walls using Microwave Oven and be as accurate as now.

      --
      Who logs in to gdm? Not I, said the duck.
    4. Re:Code source or it didn't happen by kk49 · · Score: 3, Interesting
      --
      You can have your god back when you are old enough to handle the responsibility.
    5. Re:Code source or it didn't happen by wagnerrp · · Score: 3, Interesting

      While it's nice projects like this open up such technologies to hobbyists, the article is documenting something very novel. Traditional RADAR has trouble seeing through walls, as it cannot filter out all the noise from the wall's reflection. By sending two specially encoded signals that cancel each other out against stationary objects, they've sidestepped this issue entirely, and come up with something brand new. They've developed the much wanted Sci-Fi trope of a portable motion detector.

    6. Re:Code source or it didn't happen by kk49 · · Score: 1

      Good point, I'll have to read the article.

      --
      You can have your god back when you are old enough to handle the responsibility.
    7. Re:Code source or it didn't happen by Anonymous Coward · · Score: 0

      Do you of any idea how illiterate you sound?

    8. Re:Code source or it didn't happen by Anonymous Coward · · Score: 0

      THAT backfired.

  4. NSA Use by Oysterville · · Score: 2

    So then I guess we should be looking into ways to prevent government agencies from using this against us. They've already proven to be completely incapable of trustworthiness.

    1. Re:NSA Use by Anarchduke · · Score: 5, Informative

      How is that? They are entirely trustworthy; they do what we tell them to do. We gave them every bit of authority they use to spy on us. We threw our privacy at them and cried, "Take this! Save us from the terrorists and spy upon us!"

      --
      who prays for Satan? Who in 18 centuries has had the humanity to pray for the 1 sinner that needed it most? ~Mark Twain
    2. Re:NSA Use by AHuxley · · Score: 1

      Yes US seems right up on all the tech average people use and how to track them:
      http://www.wired.com/dangerroom/2012/03/petraeus-tv-remote/
      Become a domestic “person of interest” and they will use every connected device you have.
      Welcome to a world where you have to change your notions of secrecy and enjoy every connected device been backdoor ready by design.

      --
      Domestic spying is now "Benign Information Gathering"
    3. Re:NSA Use by Anonymous Coward · · Score: 0

      They'll put them in the smart meters, and jail you for not wanting to install them.

    4. Re:NSA Use by Trepidity · · Score: 3, Interesting

      When it comes to viewing the movement of humans through walls, there have already been infrared cameras for years, which in most situations will do anything this wifi approach can do and more. The only real advantage of this wifi approach is that it's cheaper, using ubiquitous commodity hardware.

      So when it comes to government agencies, this doesn't really change the technological situation: they've already had the ability to track movement through walls for years. They're only restricted in using it to the extent that legal restraints are successful. For example in Kyllo v. United States the Supreme Court threw out a conviction that was obtained in part by using infrared cameras to look inside a home without a warrant.

    5. Re:NSA Use by citizenr · · Score: 1

      Its funny you link to a quote by David Petraeus - he was Character assassinated by very same CIA/private contractors running surveillance show as soon as he announced CIA security audit.

      --
      Who logs in to gdm? Not I, said the duck.
    6. Re: NSA Use by Anonymous Coward · · Score: 0

      You can't stop ANYONE from using this, and the government would need a warrant to search inside your home with it. The whole seeing/hearing through walls thing is not some brand new legal challenge, go look it up instead of whining on slashdot.

    7. Re:NSA Use by Anonymous Coward · · Score: 0

      To be fair, Petraeus is a total cunt and had it coming.

    8. Re:NSA Use by tibit · · Score: 1

      When it comes to viewing the movement of humans through walls, there have already been infrared cameras for years, which in most situations will do anything this wifi approach can do and more.

      Can we trade user id's? I'd expect a bit more from a 597. What you claim is an unoriginal figment of your imagination.

      --
      A successful API design takes a mixture of software design and pedagogy.
    9. Re: NSA Use by Anonymous Coward · · Score: 0

      Infrared cameras cannot see through walls. Same goes for thermal imaging. They have wavelengths very close to visible light (in the nano-meters); therefore, they cannot penetrate walls. Wi-Fi has much longer wavelengths (12.5 centimeters) and therefore has much much higher penetration.

    10. Re:NSA Use by Jane+Q.+Public · · Score: 1

      "I'd expect a bit more from a 597. What you claim is an unoriginal figment of your imagination."

      I'd expect more from someone who expects more. Your argument is irrelevant to what he was saying.

      While thermal imagers may not "see" what is inside a home, in the sense that they can't pick up shapes in any detail through a wall (the signature becomes too diffuses), they still can impart significant information about what is going on, on the other side of those walls. That is the relevant issue here.

      The point being: something that can't see much more than a warm blob is already illegal for the police to use without a warrant. So something that actually captures motion doesn't even come within a country mile of being usable by police without a warrant.

    11. Re:NSA Use by Anonymous Coward · · Score: 0

      Doesn't matter. The police aren't above using it first, then coming up with a reason like:
      "Anonymous report" - by nature can't be verified
      "Neighbors complained" - no need to verify that at all
      "High electric bill" - totally proves a grow-op
      "It smelled" - yup, they can smell through walls too...no they are just saying that because they FLIR'd you and don't wanna fess up to it.

    12. Re:NSA Use by tibit · · Score: 1

      ?!

      When it comes to viewing the movement of humans through walls, there have already been infrared cameras for years

      Infrared cameras can't see movement through walls, and they definitely can't come even into the same performance ballpark, when it comes to detection of humans, as sensing that uses longer waves. Heck, if you're a lone human in a large enough room, a thermal camera can't even tell that you're there. The surface temperature won't rise enough due to large wall area.

      --
      A successful API design takes a mixture of software design and pedagogy.
    13. Re:NSA Use by Jane+Q.+Public · · Score: 1

      "?! "

      I think your reading comprehension skills need a little work. That isn't even close to what I wrote.

    14. Re:NSA Use by Jane+Q.+Public · · Score: 1

      They tried that in NY and got busted, big time.

      Jeff Cooper, ex-Texas State Trooper, found out police were IR scanning the outside of homes in NY, looking for growing operations. They set up something... I think it was a Christmas tree, don't remember... with some grow lights, then some remote cameras.

      When the cops busted the door in, it was all caught on camera. They found a note saying "Busted!" and a few other choice things, and the video was made public.

  5. in other news by waddgodd · · Score: 1

    MIT researchers can become wet using water. Seeing through things by using wavelengths that penetrate them isn't particularly new, and one of the selling points with wi-fi is that it can go through walls Non-news for nerds, stuff that natters

    --
    Just because you're paranoid doesn't mean they aren't out to get you
    1. Re:in other news by Anonymous Coward · · Score: 0

      Yes, but if I am an uberspy, and sitting in my hotel room with my spectrum analyzer on my desk, watching for errant radar signals that may be examining my movements, WiFi frequencies are much less likely to arouse my suspicions than something in other frequency bands. So I merrily go about my spywork thinking that there's nothing going on... but drat! I'm caught by the NSA and their nefarious WiFiRadar.

      At least that's how it will work in the movies.

    2. Re:in other news by solidraven · · Score: 1, Interesting

      Yes sure, everybody has a field service unit spectrum analyser that goes all the way down to -150 dBm in their hotel rooms!!!! Fact is that there are a lot better ways to track somebody. Radar isn't exactly something new either. I've built radars out of scrap minicircuits components that I found in my desk drawers, and even without trying you could see faint movements through walls on short ranges if your receiver's noise figure is sufficiently low.

    3. Re:in other news by neonsignal · · Score: 2

      It's somewhat more difficult than just 'seeing through things'; walls are not totally transparent to the wavelengths of interest, and reflections from them dominate the signal. Moreover, this technique makes use of equipment that can't do the precise timing of radar measurements (though this means they can only track the angles of moving targets, not their position). It is a neat sort of hack, and interesting for that reason.

    4. Re:in other news by Anonymous Coward · · Score: 0

      There is a significant difference between "uberspy" and "everybody", no matter how many exclamation points you add to the sentence.

    5. Re:in other news by wagnerrp · · Score: 1

      They're actually using SDR units, rather than commodity WiFi gear, so they probably could rig it up to operate as a proper RADAR system. They just don't have the precision necessary to track someone as they move through a small room.

    6. Re:in other news by tibit · · Score: 1

      What they are doing, in fact, is a well known measurement technique. They tweak the transmitted signals so that the receiver sees a null. Then, as soon as you get moving objects introduced into the volume, the receiver get a signal that's mostly related to this object's motion. I have used the same trick with ultrasound transceivers back in high school, I wish I had it written up.

      --
      A successful API design takes a mixture of software design and pedagogy.
  6. :O by Hunterhawk · · Score: 2

    Sweet I can finally watch my neighbors T.V. And to think people used to use WINDOWS! hah! so crude.

    1. Re::O by PolygamousRanchKid+ · · Score: 1

      Sweet I can finally watch my neighbors T.V. And to think people used to use WINDOWS! hah! so crude.

      No, we use TEMPEST for that.

      --
      Schroedinger's Brexit: The UK is both in and out of the EU at the same time!
  7. Not Wi-fi by wisnoskij · · Score: 4, Informative

    I am sure they are using the same frequencies as wi-fi. But wi-fi is not just the use of a certain frequency range.

    --
    Troll is not a replacement for I disagree.
    1. Re:Not Wi-fi by Trepidity · · Score: 1

      Ah, good catch. In that case, if they aren't actually using existing wifi hardware, why pick that particular frequency range? It doesn't seem like it has particularly great properties as far as the intended application. Is it just due to regulatory issues, since it's in the mostly-unregulated ISM band?

    2. Re:Not Wi-fi by wisnoskij · · Score: 1

      They very well might be using existing hacked wi-fi hardware.

      But if I used wi-fi hardware to fry an egg, that does not mean that I fried an egg with wi-fi.

      --
      Troll is not a replacement for I disagree.
    3. Re:Not Wi-fi by dissy · · Score: 2

      It looks like they are using standard wifi hardware, two modules actually.
      What they aren't doing is speaking 802.11 over it and are providing their own framing in order to get a reversed signal on the other module. There will be no protocol in that framing to speak IP on top of.

      The major advantage is that a manufacturer can purchase the same wifi chips as they usually do for wifi, but these two chips are configured in software quite differently. I assume they will likely have a different antenna to optimize it for this function too, since the hardware modules will be dedicated to this task.

      That's also probably why they don't call this "wifi" at all. (Go go gadget slashdot reporting!)

      While it might be possible in some existing devices to reconfigure the existing wifi chip to work in this way, you must have two in total for the task or it won't work, so most existing devices will be excluded due to that reason alone. Of course any antenna issues will make it sub-par, and also you won't be able to use a chip for both this and wifi at the same time which would be very handy for a remote sensor type device.
      Most likely, any new hardware built for this purpose will come with a total of 3 hardware wifi modules, one dedicated to 802.11 and the other two dedicated to wivi.

      Using existing mass-produced hardware will let maufacturers take advantage of both the low cost of those modules, and the fact they likely have them in their ordering system with bulk pricing deals setup already.
      Previously one needed dedicated sonar or radar hardware, which isn't in as large of demand and so still pretty expensive.

    4. Re:Not Wi-fi by Ion+Berkley · · Score: 1

      You are correct in the sense that they are not using the upper layer protocols in Wifi, but incorrect in that the signal modulation they are using is that of modern Wifi (OFDM) as is there intended channel width (20MHz) and frequency band (2.4GHz ISM). So basically they are using a flexible R&D H/W solution (The USRP) to transmit and receive a WiFi like signal and demonstrate their innovation. Existing WiFi IC's would not be able to easily do this because they are not designed for new experimentation but just to implement the standard as cheaply as possible. The point being when you have proven, demonstrated and refined the concept using a prototyping platform, then you look for manufacturer(s) to incorporate the technology in production silicon...they are not close to that point yet but have just demonstrated an important milestone.

    5. Re:Not Wi-fi by tibit · · Score: 1

      Never mind that at least the transmitters must be running from same reference oscillator, if not all three devices (two transmitters and the receiver).

      --
      A successful API design takes a mixture of software design and pedagogy.
  8. Cop-out by Dcnjoe60 · · Score: 2

    What a cop-out! From the article "Like all technologies in the world, it depends on us how we use them," Katabi said.

    People use that excuse all the time. However, when one has a reasonable expectation as to how some technology will be used, you cannot fall back on this reasoning to ease one's conscious over their own culbability in something. The scientists involved in the Manhatten project new full well how that technology would be used and had to deal with the moral implications. Today, though, the notion is technology for technology's sake with no thought of the consequences.

    Well here's news to all my fellow researchers, if you develop some new technology, great, but don't hide behind your mommy's skirts and say, something like the above. You know darn well how it will be used, so take responsibility for it.

    1. Re:Cop-out by Anonymous Coward · · Score: 0

      MIT has always been a technocrat's paradise, and an ethicist's nightmare.

      But you could say that about all of America, and now the Sino-American conglomerate, ever since Vannevar.

    2. Re:Cop-out by DigiShaman · · Score: 1

      Any technology that has been invented, eventually would have been (because we now know it's possible). While not a direct quote, it's what Edward Teller said more or less of the Hydrogen Bomb. That, and we would have been speaking Russian now had we (the US) not pursued the H-Bomb program.

      --
      Life is not for the lazy.
    3. Re:Cop-out by Dcnjoe60 · · Score: 2

      Any technology that has been invented, eventually would have been (because we now know it's possible). While not a direct quote, it's what Edward Teller said more or less of the Hydrogen Bomb. That, and we would have been speaking Russian now had we (the US) not pursued the H-Bomb program.

      That is true, but some technologies are much more readily apparent as to their misuse than others. Splitting the atom is a good example. It is hard to think of peaceful applications of such a technology (which is different from, say, nuclear power). In the past 40 years or so, science has focussed more and more on can we do such and such and ignored the question of should we be doing such and such.

      However, we scientists can not hide behind the moral argument that we are only inventing technology it isn't up to us how it is used, any more than the Nazi solders could claim they were only following orders. Scientific discovery has the power for good and evil and to ignore the potential of evil, particularly if it far outweighs the good is either a sign of moral weakness or culpability.

      This doesn't just apply to scientific research, but all forms of engineering and product development in our society. I am quite confident that we could put a 500hp engine in a vehicle the size of a Fiat 500. The capability is there, but should we do that? Such an overpowered vehicle would be a death trap and the only thing that keeps a manufacturer from doing so is the liability claims would outweigh the profits from such a vehicle. Auto manufacturers aren't allowed to hide behind we only build the car, its up to people how they use it (although gun manufactures do use that thought pattern). But, for scientific research, it is acceptable.

      MIT developed a new way to track people inside buildings that the only really useful and practical application is for surveillance. There is nothing wrong with that, until they come out downplaying that use and trying to say how it will actually help make privacy even better (in the article). Knowing that is a losing battle, and evidently feeling ashamed of their discovery, they then fall back on the old we only invented it, it's not up to us how people use it.

      If Snowden had released that the NSA was already doing this, people would be outraged. But then MIT isn't responsible for how the government uses the surveillance technology they develop. Maybe they should require all those MIT students to take some basic philosophy and ethics courses. It would be helpful if the brightest minds in the country could tell right from wrong and how to take responsibility for what they are doing.

    4. Re:Cop-out by Anonymous Coward · · Score: 0

      Re: " It is hard to think of peaceful applications of such a technology (which is different from, say, nuclear power)."

      You are aware, are you not, that fissioning atoms are the fundamental technology behind both nuclear bombs and nuclear reactors? The first takes place in a fraction of a second and the second takes place over 20 years. There's a lot of container and associated tech that's different too... but the fundamental fissioning process is exactly the same.

  9. Told ya so by sjames · · Score: 1

    I tried to tell you people those newfangled WiFi thingamagiggies would mutate your DNA, but did you listen?

    Now you got mutants at MIT seeing through walls and who knows what else.

  10. CSAIL once again invents useless technology by Anonymous Coward · · Score: 0

    It only works well without numerous other wi-fi devices obscuring the signal. They share department space with Robert T. Morris, which explains how they their department gets funding for useless work. (Robert Tappan Morris, the author of the "Morris Worm" that helped effectively take down much of the Internet in 1988, and whose father as head of the NSA helped prevent his serving a day in jail.)

    It's where you *hide* MIT computer professors who don't do actually useful work, but you want to keep them around for political reasons.

    1. Re:CSAIL once again invents useless technology by tibit · · Score: 1

      I agree that while the idea may be novel, its benefits are useless in actual emergencies. When there's an emergency, no one gives a flying fuck about whether your radar might cause interference. Blasting a wideband signal is fine, especially that wideband is resistant to narrowband interference (the usual kind from intentional emitters). Just imagine the backlash at FCC if the police was using an interfering radar during a hostage situation (even better if kids are involved) and someone at FCC had the gall to raise concern. Instant career ender.

      --
      A successful API design takes a mixture of software design and pedagogy.
  11. Just like T-Rex by Megahard · · Score: 4, Funny

    It can't see you if you don't move.

    --
    I eat only the real part of complex carbohydrates.
    1. Re:Just like T-Rex by Anonymous Coward · · Score: 0

      The signal passes through things. Good luck having a still, beating heart. I heard the last guy to figure it out is in jail.

  12. Freq? by behrooz0az · · Score: 1

    We already have enough congestion on 2.412~2.482.Why on earth did they choose Wi-dear lord-Fi?

    --
    Moderating "-1, Disagree" is simple censorship. Have the guts to post your opinion. -- Spazmania (174582)
    1. Re:Freq? by wagnerrp · · Score: 1

      Small enough to actually track people, large enough to penetrate walls, and free to use within limited output power.

  13. Where's my.... by Pav · · Score: 1

    ...functional Aliens motion tracker???

  14. There's an inconsistency in the demo. by tibit · · Score: 1

    There's an inconsistency between what we see in the demo and the description. Supposedly the system tracks the angle of the object. Yet what the graph in the video looks like is nothing like the angle. It looks like a simple Doppler output that goes to zero once the subject stops moving.

    --
    A successful API design takes a mixture of software design and pedagogy.
    1. Re:There's an inconsistency in the demo. by tibit · · Score: 1

      Never mind that their project website is a piece of crap written for the illiterate. Give me a fucking break, it sounds like a something a particularly clueless high-schooler would write upon seeing the project.

      --
      A successful API design takes a mixture of software design and pedagogy.
  15. See through walls not new by Anonymous Coward · · Score: 0

    High tech ways aren't always better. Heck, the old reliable "dill a hole" in the wall then look through lets one see both moving and non-moving things. Only problem is those being watched sometimes get upset. Suppose that's why the researchers turned to high-tech ?

  16. WiVi's nemesis? by Anonymous Coward · · Score: 0

    WiVi meet .... roomba

  17. Tech Progression by Anonymous Coward · · Score: 0

    It's interesting to see how tech progresses. We're starting to get the technology to actually see through or around things and we're also starting to get the technology to make things invisible. Both are mostly in the scientific stages and both have working demos. I hope they make it to the consumer stages at the same time, though I expect seeing through things to be first as it appears to be cheaper and easier than making them invisible.

  18. "Wo-Vi can translate...reflections...into... perso by Anonymous Coward · · Score: 0

    Wi-Vi can only translate reflections into displays of reflectors' movements. Put the reflectors on a stick, a mobile, a wind-chime or a yo-yo and you can watch the movements of that, while the persons do whatever they want. What you need are infra-red detectors to track the people, until they don Mylar heat reflector suits, when you should be able to track them with your Wi-Vi -- Unless they put overcoats on over their Mylar reflector-suits, when you will only see them when they flash you.

  19. Block it? by Anonymous Coward · · Score: 0

    So it would seem that simply sheeting walls with sheets of tin foil would spoil everything.

  20. Already seen it done & better with multiple st by Beylerian · · Score: 1

    This research achieves the same and even better (learns and detects multiple states) with an even more interesting approach. http://search.ieice.org/bin/summary.php?id=e95-b_10_3088&category=B&year=2012&lang=E&abst=

  21. I can see through walls too.. by vedranius · · Score: 0

    by using windows, hehe.