Forgot your password?
typodupeerror
Input Devices Technology

MIT Develops "Kinect of the Future" 76

Posted by samzenpus
from the all-the-better-to-see-you-with dept.
itwbennett writes "Using radio signals, MIT researchers can pinpoint someone's location — through a wall — with accuracy of +/- 10 centimeters. Fadel Adib, a Ph.D student on the project, said that gaming could be one use for the technology, but that localization is also very important. He said that Wi-Fi localization, or determining someone's position based on Wi-Fi, typically requires the user to hold a transmitter, like a smartphone for example. 'What we're doing here is localization through a wall without requiring you to hold any transmitter or receiver [and] simply by using reflections off a human body,' he said. 'What is impressive is that our accuracy is higher than even state of the art Wi-Fi localization.'"
This discussion has been archived. No new comments can be posted.

MIT Develops "Kinect of the Future"

Comments Filter:
  • this will be suppressed and classified in 1, 2, 3...
    • by mcgrew (92797) *

      Yeah, getting that genie back in the bottle will surely be successful. Right.

      • by icebike (68054)

        Well, the "bottle" in this case requires "three radio antennas spaced about a meter apart and pointed at a wall and A desk full wires and circuits to generate and interpret the radio waves.".

        So keep an eye out for those things appearing in your home, and be forewarned that suddenly appearing antennas and a large amount of circuitry might indicate someone wants to know exactly when you are in front of the TV. When its small enough to be built into the TV, you might not notice it, but a camera would be mor

      • Classification doesn't need to be successful: there are people whose job it is to classify things and to show how hard they are working, they have to classify a lot of things. If your work is for a military use, sometimes things as unclassifiable as printouts from Wikipedia that you have used may have to be locked up in classified storage with big red "CLASSIFIED" stamps on them. Then any work that used that information has to be classified as well.

        I don't know about any jobs that involve the suppression of

    • Why suppress and classify? Encourage everyone to have one in the home and have a backdoor installed. That would fit with the usual NSA modus operandi.
    • This was classified in 1,2,3 ... in 1941, 1942, 1943.
      Since then, radar has been declassified.

      Slashdot editors: Please add a few lines in the summary to trigger interest for futue inserts.
      Currently this one just says "researchers at MIT built a radar" - and what is the news in that?
      In the editors' defence the article doesn't say that much more, either.

    • by fractoid (1076465)
      You mean it will be suppressed and classified in 1996 and will only be released to the public in 2013, two years after passive shielding is developed and the first soldiers are equipped with active scanners to pinpoint "through-the-wall kinect" sensors from 200m away?
  • by Barbarian (9467) on Sunday October 13, 2013 @11:46AM (#45114635)

    I am sure if you can easily pinpoint humans through a wall with 10 cm accuracy, you will see this funded and developed by your favourite defense contractors. Sniper rifle with wallhack, anyone?

    • by moschner (3003611)

      My first thought was to use it in search and rescue. It would be very useful to be able to detect people trapped under rubble or inside a burning building. Send in a small robot with its own wifi and find out where people are trapped. With a little refinement, maybe it could be built into first responder gear or firefighter helmets.

      • It would be very useful to be able to detect people trapped under rubble

        I'd assume that one of the features is that it detects people by means of detecting *changes* in the signal, not by magically reconstructing a 3D picture of anything static from a few waveforms. People trapped under rubble sort of tend to not walk around in open space a lot.

        • by Wescotte (732385)

          Maybe flailing your limbs around is asking too much. However, if we're lucky just the acting of breathing itself will produce enough of a change to detect a person?

    • Just what I was thinking. Coupled with AR-googles, you've got "x-ray vision" of sorts. Better still if sonar is giving density cues on the walls between you & your target(s)...
  • by TechyImmigrant (175943) on Sunday October 13, 2013 @11:49AM (#45114651) Journal

    >Using radio signals, MIT researchers can pinpoint someone's location — through a wall — with accuracy of +/- 10 centimeters. Fadel Adib, a Ph.D student on the project, said that gaming could be one use for the technology,

    I suspect killing people will be higher on the list of priorities of certain funding bodies.
     

    • by Ksevio (865461)
      But we already have tools that do that - being able to put that in the living room is the step forward.
  • by Anonymous Coward on Sunday October 13, 2013 @11:58AM (#45114677)

    Yes, people have been doing "radar through walls" with all manner of signals for decades.
    Here's the problem.. works fine on a simple homogenous wall: the wall looks like a sheet of glass to the radar.

    Now, what if that wall is more like pebble or frosted glass? Or say, glass blocks (which is what standard concrete blocks or bricks look like at microwave frequencies.

    What if the house were a "hall of mirrors" since the walls can reflect microwave signals, and some goes through. What about furniture?

    Sure, one *could* with enough measurements and computation figure all this stuff out (it's called an inversion problem), and then calculate what's actually going on.

    Detecting the reflections is the easy part. We've all done that with over the air TV back in rabbit ear days. Turning that into an image is substantially more challenging.

  • by RyanFenton (230700) on Sunday October 13, 2013 @12:08PM (#45114731)

    Using radio waves to track position/movement is VERY far from new. Even imaging through walls is extremely old hat. A control interface with 10cm+/- resolution would be drastically worse than any current game controller.

    This isn't for gaming - it's use would be primarily for surveillance and automated 'security' tools of various kinds. It's not Xbox - its NSA/military/creepy 'spy' tools.

    Ryan Fenton

  • by Kwyj1b0 (2757125) on Sunday October 13, 2013 @12:18PM (#45114791)

    As a counter measure, I was wondering: can this be stopped by designing the right wall?

    If you make a wall that scatters a lot back due to a non-homogenous constitution or imperfections, will there be enough reflections to make the system useless? And is that a universal flaw (i.e no matter what type of radar you build, a single "well designed" wall can thwart all such systems)? Or does the wall need to be designed specific to the frequency/design of the detector?

    The reason I ask is that I consider this is likely to be an invasion of privacy - "we didn't need a warrant to track the subject, the neighbor was a perv. who had the scanner, and came forward (and we aren't filing charges against him for $unrelated_crime)." It would be nice if a low-tech solution can thwart this.

    • The traditional approach for jamming radar is lots of small slivers of metal foil. In this case nails hammered into the timber frame might do the trick. Maybe even electrical wiring would work (esp. if you carry data over it, too).

      I bet this radar is not too clever when metalised insulation panels are used in the wall cavities, either.

    • by Anonymous Coward

      Given that with today's technology, it would be tough to do realistic "through the wall imaging", especially with a WiFi as the illuminator. Google reports from the Army Research Lab on "Through the Wall Radar" for some real data.
      Let's assume, though, that there's enough processing power to solve the substantial problems of making an imager using WiFi as a practical fieldable device (as opposed to a mass of wires on a desk, and tracking one moving target in the empty next room through a thin wall).

      But yes,

      • by PPH (736903)

        But you may never know when or where you'll need to jam/confuse such radar. Better to have some inflatable Mylar balloons of a roughly human shape and size. When you find yourself in a room and at risk of detection, you inflate them and let them drift around you in the room. Too many targets confuses the radar operator/assassin.

        If you shield one room, your attacker will be able to identify your hiding place. Or the location of activities you don't want observed.

    • Stealth technology merging with home design, lots of curved surfaces and irregular density walls in the ideal home of the future.
    • by fishicist (777318)

      Imaging through scattering media, for noble causes such as medical imaging, is a current and productive research area in optics. You can account for essentially arbitrary scattering as long as it doesn't change quickly. While the technology would be different, but I expect that the mathematics and the techniques already exist to thwart and such wall which you might design.

      On the other hand, a 5mm sheet of aluminium ought to do the trick. :)

    • Google "foil wallpaper". It's electrically conductive and once grounded you have a Faraday cage.

    • by dix719 (963933)

      The reason I ask is that I consider this is likely to be an invasion of privacy - "we didn't need a warrant to track the subject, the neighbor was a perv. who had the scanner, and came forward (and we aren't filing charges against him for $unrelated_crime)." It would be nice if a low-tech solution can thwart this.

      This sounds like a job for the "old-school" Aluminum Foil hats...

    • by KVM (2949923)
      Well a faraday cage for starters could block all radio frequencies, and a mesh cage like the thing that line up microwave door could selectively block some wavelengths, like microwave door allows visible light to pass but doesn't allow microwave to pass. And the faraday cage could be a nice metal foil or mesh sticked to the wall
  • "Shown publicly this week for the first time the project from MIT's Computer Science and Artificial Laboratory (CSAIL) "

    I think the most impressive part is not that they can do it, but that they developed it in an artificial laboratory! (i.e. Should be Artificial Intelligence laboratory.) So basically, while the MIT guys can detect things trough walls, your typicall IT World editor can't get an extremely small article right. Worse yet, the typo has been there since October 11, 2013, 12:14 PM. Really IT

  • It's really sad that even though I really had no interest in this topic, I watched the video just because the video preview clearly showed they were using Ubuntu. Yet, disappointingly, they didn't cover anything about the control interface they were using; this could be used as a push for gaming on Linux desktops, especially since they don't have plans to make money off of it, and could hence afford to have a slow start.
  • +/- 10 cm resolution is not in any way useful as a "Kinect of the future".
    Why did this become a story on slashdot? It's total crap.

    • by thejynxed (831517)

      I take it you missed the part where that +/- 10 cm was THROUGH A WALL.

      • by gl4ss (559668)

        ..yeah? since when is kinect supposed to work through a wall?

        this is this years update on what they did last year at the same place. they couldn't find any real use for it so they slapped a game on it.

      • by Gibgezr (2025238)

        Are you implying that, where the wall not there, this would then qualify as a "Kinect of the future"?

  • by Anonymous Coward

    I wonder if the people researching this stuff ever think of the real life implications of the technology. Do they really want a world where they have no privacy even inside their own home or bedroom? not to mention all the people this tech will enable military and police to more easily kill.

    I wish I could have a reaction of "oh wow that's pretty cool" but instead it's just depressing. I don't want to go back to the dark ages but I no longer really get enthusiastic with the way technology is moving these da

  • Something with reclining leather seats, that goes really fast, and gets really shitty gas mileage.

  • interesting......

Every nonzero finite dimensional inner product space has an orthonormal basis. It makes sense, when you don't think about it.

Working...