Slashdot Mirror

← Back to Stories (view on slashdot.org)

Smartphones Can Steal 3D Printing Plans By Listening To The Printer (fedscoop.com)

Posted by BeauHD on from the always-listening dept.

An anonymous reader quotes a report from FedScoop: Smartphones equipped with special programming can become a sophisticated spy sensor capable of stealing designs from a 3D printer -- just by measuring the noise and electromagnetic radiation the printer emits. Researchers from the University of Buffalo recently discovered how a smartphone on a bench about 8 inches away from a 3D printer could allow someone to reconstruct a simple object being printed with 94 percent accuracy. Complex objects can be copied with 90 percent accuracy. The attack basically reverse-engineers the printing blueprint by reconstructing the movement of the nozzle from the electromagnetic and acoustic energy it generates while working. Most information came from electromagnetic waves, which accounted for about 80 percent of the useful data. The remaining 20 percent came from acoustic waves. Wenyao Xu, assistant professor in the University of Buffalo's Department of Computer Science and Engineering, is the lead author of the study, "My Smartphone Knows What You Print: Exploring Smartphone-Based Side-Channel Attacks Against 3D Printers," which will be presented at the Association for Computing Machinery's 23rd annual Conference on Computer and Communications Security next month in Austria.

45 comments

  1. More "research" by 110010001000 · · Score: 4, Insightful

    Yes we get it: you can obtain information by listening and measuring things. I can write software that can identify a car and how fast it is travelling with 90% accuracy just using the microphone on a smartphone. Is this what passes for research today in CS? If so, we are doomed.

    1. Re:More "research" by friesofdoom · · Score: 2

      Mod this guy up plz!

      Next up: Researcher discovers he can navigate in the dark using touch!

      Just wait till they figure out braille!

    2. Re:More "research" by 110010001000 · · Score: 0, Troll

      No! Mod him down! He is a known crackpot!

    3. Re:More "research" by networkBoy · · Score: 1

      Actually it is a good bit of research.

      Conceptually, sure it was straightforward. That doesn't mean the implementation was simple. A huge part of stuff like this is what you learn while doing it. Sometimes things that seem obvious turn out not to be.

      A famous example:
      Supersonic flight. At some point as you cross the threshold your up and down are reversed in supersonic flight. Early pilots pulled up to avoid the ground and as a result crashed harder. Pretty counter intuitive, but that was the case.

      I would bet that some of the things that intuitively went together (like improving the accuracy of the EM wave interpretation with results from the sonic detection) had lots of gotchas in it.

      -nb

      --
      whois gawk date unzip strip find touch finger mount join nice man top fsck grep eject more yes exit umount sleep dump
    4. Re:More "research" by paradigm82 · · Score: 1

      Completely agree - seems there's so much of all this rather obvious trivial research. Especially in the CS security area, where everyone tries to find some new angle or some new "attack" even though it really isn't a violation of any security propreties (as in this case). The more useless and theoretical it is, the bigger the fanfare. It is the same in other areas - very simple questions/discoveries (i.e. just something from questionaires etc.) being hailed as big breakthroughs. Guess it is the combination of the extreme influx of PhD's students in recent years combined with the desire/expectation/requirement of constant (media) attention for every simple thing etc.

    5. Re:More "research" by Anonymous Coward · · Score: 1

      These types of attacks are well known, have been done in real life for decades, and can be done against anything that emits anything while operating. The only thing that's different is they used a cell phone instead of dedicated listening hardware. Since cell phones are more common and are packed with sensors, there's really nothing interesting in that.

      Currently this is a really dumb attack, though once they improve the sensors and are able to get the software working through walls it'll be slightly useful (other similar attacks already do that). For this to work the phone has to sit there until the print is done. It'd be far easier to snap a couple pictures of the item after it's been printed then stitch those images together for a 3D model. There's already consumer software that does that and it only takes seconds instead of hours and hours.

    6. Re:More "research" by RabidReindeer · · Score: 2

      This "obviously trivial" research is worth a lot of money to the right people.

      Just because you "know" something can be done doesn't mean that it can be done easily, economically, or effectively. And then there's the stuff that people "know" that's not so at all.

      People in industrial and military espionage can now take heart in the knowledge that all sorts of top-secret plans can be stolen with 90+% accuracy by simply "accidentally" leaving a phone on a workbench. That means that they don't have to chance possibly less unobtrusive ways that they might also "know".

      Conversely, security-minded installations have one more reason to prohibit cellphones in sensitive areas and even not-so-obviously sensitive ones like open production floors.

      Now, what do we "know" about that FitBit you're wearing?

    7. Re:More "research" by JustAnotherOldGuy · · Score: 1

      No! Mod him down! He is a known crackpot!

      Lol, you may be a crackpot, but your original post was spot-on.

      The only "newsworthiness" of this article is that it involves 3D printing. Otherwise, it's a non-story. Actually, it's a non-story either way.

      --
      Just cruising through this digital world at 33 1/3 rpm...
    8. Re:More "research" by Obfuscant · · Score: 1

      People in industrial and military espionage can now take heart in the knowledge that all sorts of top-secret plans can be stolen with 90+% accuracy by simply "accidentally" leaving a phone on a workbench.

      If a 3D printer is printing a top secret thing, don't you imagine that the people running the printer might know that any phone "accidentally" left on a workbench is a problem? Like, "we're in a top secret facility working with top secret material -- how did ANY smartphone get in here?"

      and even not-so-obviously sensitive ones like open production floors.

      Top secret parts are not going to be printed on an "open production floor".

      On the other hand, maybe the next Secretary of State will have a flunky email the top secret CAD files to her home server so they can be printed on the home 3D printer. Wait, no, that's not a problem. If the printed part doesn't have a coversheet declaring it Top Secret it obviously cannot be. Nevermind.

    9. Re:More "research" by gordguide · · Score: 1

      Yes, some /.'ers may not remember this one, I think it dates back to the mid 1990's, but it was determined you could accurately discover every keystroke on a keyboard and correctly identify the keys struck, by using a microphone, including built-in microphones in PCs and multimedia devices, or hands-free sets. It did not require the use of "loud" keyboards using the old IBM/Apple style scissor key mechanisms. Certainly a smartphone app would do the job nicely today.

    10. Re:More "research" by Anne+Thwacks · · Score: 1
      More to the point, if they are chucking out this amount of RFI, it is hard to see how they comply with European EMC requirements.

      Some people need to learn about screening and earthing.

      --
      Sent from my ASR33 using ASCII
    11. Re:More "research" by smallfries · · Score: 1

      In related news: somebody with physical access to the printer can also photograph the object - or even take it!

      --
      Slashdot: where don knuth is an idiot because he cant grasp the awesome power of php
    12. Re:More "research" by drinkypoo · · Score: 1

      Some people need to learn about screening and earthing.

      Most 3d printers have none of the former and precious little of the latter, since they're mostly made of plastic

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    13. Re:More "research" by RabidReindeer · · Score: 1

      There are secrets and then there are Secrets. Sometimes things are done casually that wouldn't be done if there were explicitly-known attack vectors. Sometimes an "accident" might be forgivable if it was truly thought to be innocuous.

      This has nothing to do with specific persons or political parties. If you insist on sliming the topic with drive-by political sniping, I'll have to ask Trump to sic our state officials on you.

    14. Re:More "research" by LifesABeach · · Score: 1

      So, is Reverse Engineering something new? And for all those not considering 3D Printing, what is Plan B?

    15. Re:More "research" by WallyL · · Score: 1

      Or if somebody has hijacked your phone and can use its features without your knowledge, they can get your 3D printer plans too, no pointing your phone at the printer required!

    16. Re:More "research" by tlhIngan · · Score: 1

      If a 3D printer is printing a top secret thing, don't you imagine that the people running the printer might know that any phone "accidentally" left on a workbench is a problem? Like, "we're in a top secret facility working with top secret material -- how did ANY smartphone get in here?"

      Who said the stuff had to be top secret? Theres tons of 3D printed stuff that's not top secret, but plenty confidential. Automakers go to great lengths to hide the body shape of new models - cloaking them with body panels from other cars in their line. I'm sure in the design room things are a lot more casual and when people are prototyping models, they use a 3D printer a lot. Even if other automakers aren't interested, plenty of press and blogs would LOVE to get their hands on advance sneak peeks.

      And there are plenty more industries where 3D printing is prevalent and people take a more casual stance to technology. Hell, many university research labs are ideal for this sort of industrial espionage - there's lots of people filtering through them, and a smartphone left nearby wouldn't raise any attention at all.

      Top secret parts are not going to be printed on an "open production floor".

      True, but many regular parts are While a lot of it is probably patented and thus easy to figure out how it works, obtaining the original manufacturing might be helpful if you want to know how something is manufactured. Given how slow 3D printing is, when it's used in a production line, it has to be very efficient, so even learning how the production line is made efficient can be a valuable trade secret.

      More to the point, if they are chucking out this amount of RFI, it is hard to see how they comply with European EMC requirements.

      Some people need to learn about screening and earthing.

      There's lots of RFI you can get - you just can't exceed certain limits. Even a properly shielded and screened unit will emit RFI. Hell, cables are not generally tested for RFI - and they're usually the largest source of emissions (often times as it's only the unit being tested, the test will use shielded cables to prevent external RFI from being part of the measurement. Well, it also means RFI conducted out of the unit through regular cables can emit quite a lot). And yes, I've gone through the RFI procedure. Our product took a 1080p input through HDMI, and ensuring our 148.5MHz (pixel clock of HDMI at 1080p) emissions were not caused by the source device (Blu-Ray player) or cables was not exactly easy.

  2. I can't wait for the future! by dohzer · · Score: 2

    Anyone else want to fly on a cheap knockoff plane who's fuselage is 94% correct?

    1. Re:I can't wait for the future! by Obfuscant · · Score: 1

      Anyone else want to fly on a cheap knockoff plane who's fuselage is 94% correct?

      Anyone want to fly in a plane that is built on a 3D printer, even 100% correct?

      Really, if you can get your phone next to the printer to measure the electromagnetic energy it emits, why not just have the phone measure the electromagnetic energy that reflects off the printer using the ubiquitous spy software called "camera"?

    2. Re:I can't wait for the future! by frnic · · Score: 1

      Well, the good news is that like all other technology, it will never get better, so we don't have to worry about the theft threat.

    3. Re:I can't wait for the future! by WillAffleckUW · · Score: 2

      Actually, some of the parts of that plane you use are, in fact, built on a 3D printer.

      So, to quote someone "you're soaking in it".

      Or, to the point, you're flying in a plane built partially with 3D printed parts.

      Keep up, it's 2016, not 1996.

      --
      -- Tigger warning: This post may contain tiggers! --
    4. Re:I can't wait for the future! by Obfuscant · · Score: 2

      Actually, some of the parts of that plane you use are, in fact, built on a 3D printer.

      "Some parts" is hardly what was meant by either the OP or me. But I think you knew that.

      Or, to the point, you're flying in a plane built partially with 3D printed parts.

      Having a bezel on the seatback display printed in a 3D printer is a lot different than having a 3D printed fuselage. But I think you knew that. You're not "to the point", you are ignoring the point.

    5. Re:I can't wait for the future! by thinkwaitfast · · Score: 1

      The majority of the parts were fabricated on CNC and EDM machines. The technology there is not stagnant. 3D printing has been a thing since about 1977. The first machines used laser sintering of plastic. Manufacturing technology is pretty interesting and actually rapidly progressing. Where are these stories other than 3D printing.

    6. Re:I can't wait for the future! by Anonymous Coward · · Score: 0

      In 1996, this this was already 30 years old. 3D printing is hardly the revolution the geeks are making it out to be.

    7. Re:I can't wait for the future! by Anonymous Coward · · Score: 0

      That's right, that's why the 747 which had its maiden flight in 1969, takes 20000 gallons of jet fuel to cross the Atlantic in six hours, but these days you can do it in ... 6 hours.

      Magic.

    8. Re:I can't wait for the future! by Anonymous Coward · · Score: 0

      or just steal the plans from the terminal.

      Which are no doubt FURTHER away than 8 inches! upgrade!

    9. Re:I can't wait for the future! by Anonymous Coward · · Score: 0

      The Jigs are 3D Printed. Not necessarily the Parts.

      But There are 3D printed parts in some plane components.

    10. Re:I can't wait for the future! by RabidReindeer · · Score: 2

      What made it a revolution was that now the parts and techniques required are cheap and ubiquitous.

      In 1989 there was a big to-do about 3-D output for computers, but the ultr-cheap miniature haptic sensors, color display units and other amenities that make stuff like the Oculus Rift possible weren't there yet.

      Even television didn't spring up out of nothing.

      Sometimes the revolution doesn't come until the time is right.

    11. Re:I can't wait for the future! by kimvette · · Score: 1

      And, some of the airplanes we fly in today are made of cloth, like fuselages were back in ~1910-~1935.

      However they aren't direct comparisons. The paper used in those fuselages back then were stiffened using a dope not entirely dissimilar to a thick shellac. Today we do the same, only we use epoxy-resins and layer the cloth.

      That 3d printing wasn't up to task back in the '70s or even just ten years ago doesn't mean that laser-sintered 3D-printed metals today are not up to task.

      Kind of like how emission controls choked performance out of cars from 1973 through 1987 or so, such that some of today's econoboxes are quicker and faster than real sportscars from the era where we hadn't learned to have good power AND clean exhaust. I wouldn't gut a cat in a new Z06 Corvette, but that's one of the first things I'd do in a C4 or L83 or L98-equipped C4 Corvette to get more power out of it (but more realistically I'd change out the engine for an LS9 or new LT4, or I'd put in high-flow/high-surface-area Random cats on a ported&extrude-honed, turbo/supercharged and tuned L98).

      --
      The Christian Right is Neither (Christian nor right). See: Matthew 23, Matthew 25, Ezekiel 16:48-50
    12. Re:I can't wait for the future! by Anonymous Coward · · Score: 0

      3d printing is for more than knobs, bezels. this is a video of a Boeing wing piece being built: http://www.digitaltrends.com/cool-tech/3d-printing-world-record/

  3. Stop giving away our intel by WillAffleckUW · · Score: 1

    Seriously, next thing I know you'll let N Korea know we have defences against them

    --
    -- Tigger warning: This post may contain tiggers! --
  4. So what? by Anonymous Coward · · Score: 0

    I already have a Yoda coffee cup.

  5. Or Just take a picture of the Object by Anonymous Coward · · Score: 0

    Or Just take a picture of the Object with the Smartphone.

  6. WOW! by thinkwaitfast · · Score: 2
    Smartphones do things a tape recorder and a computer could a decade ago. I wonder if there's an app for that.

    People were figuring out what was typed on a typewriter at least back as far as the 80s. I guess everything now done on a smartphone is new again. I can't wait for smart contact lenses can steal 3d printing plans by listening to the printer.

  7. From 8 inches away? by Anonymous Coward · · Score: 0

    Wow... It can steal the design and have a 94% accuracy rate of making another 3D printer recreate it or just think about designing it?

    If you're not concerned about people getting 8 inches away, you're probably not concerned about security at all at that point ... but even so, wouldn't it be a lot easier to just ... Oh, I dunno... grab the 3D printed object and run away? I would think at 8 inches it's possible to do so.

  8. One word... by Lead+Butthead · · Score: 1

    Tricorder. We're getting there.

    --
    ELOI, ELOI, LAMA SABACHTHANI!?
    1. Re:One word... by wonkey_monkey · · Score: 1

      Widdly-weep. Widdly-weep. Widdly-weep.

      Captain, this entire planet is made of widdly-weep.

      --
      systemd is Roko's Basilisk.
  9. Damn it! by Anonymous Coward · · Score: 0

    I worked hard on that design for the small plastic cup!

  10. Smartphone! Everything is new again! by Anonymous Coward · · Score: 0

    Those who don't know history are doomed to think they created something new.

  11. Holy shit! by Anonymous Coward · · Score: 0

    Smartphones have the ability to not only destroy the original, but put it somewhere else out of reach of the owner?

    What the fuck is this, Star Trek? We have transporter tech now?

    Or did you mean "copy the blueprints"?

    This is slashdot. Use the right fucking word, dumbass.

  12. ArqueoBit by ArqueoBit · · Score: 0

    Consultora Arqueológica ArqueoBit: http://arqueobit.com/ : plan de rescate arqueológico Perú, proyectos de investigación arqueológica, plan de Monitoreo arqueológico, certificado de inexistencia de restos arqueológicos (CIRA) Perú, plan de monitoreo arqueológico Perú, turismo arqueológico y consultoría arqueológica, Consultoría Ambiental.

  13. One Layer at a time. by Anonymous Coward · · Score: 0

    3D printers are slow. I have had some prints take longer than 24 hours.
    Who wants to leave their phone in earshot of a 3d printer long enough to hear a complete print.

  14. Two subjects by Anonymous Coward · · Score: 0

    One is reverse engineering using sound. The other is copying using a phone, which might be way easier by taking pictures.

  15. Waste of time by TJHook3r · · Score: 1

    Why would somebody go to the trouble of doing this? If you are close enough to a machine then just look at what's being printed. Or pay someone to give you the plans. Or wait for the company to base their manufacturing in a third world company where IP is not respected.