Become a fan of Slashdot on Facebook

 



Forgot your password?
typodupeerror
×
The Military

Meet Drone Shield, an Ambitious Idea For a $70 Drone Detection System 159

An anonymous reader writes "Here's an Interesting idea of how to use a Raspberry Pi and a few other inexpensive items to make a low cost detection system. From the article: 'The Drone Shield would combine a Raspberry Pi, a signal processor, a microphone, and analysis software to scan for specific audio signatures and compare them against what known drones sound like. (Because obviously a Predator drone is going to sound very different than a small quadcopter.) Once a match is found, the Drone Shield then sends an e-mail or SMS to its owner...'"
This discussion has been archived. No new comments can be posted.

Meet Drone Shield, an Ambitious Idea For a $70 Drone Detection System

Comments Filter:
  • Good job. This tech could be used for a lot of other good things as well. Basically an electronic ear for specific frequencies. I'm surprised it's not much more developed.

    Question is, will it have better acuity then a trained dog with excellent hearing?

    • Re: (Score:2, Informative)

      by Anonymous Coward

      Ofcourse it's been done. There's a lot of military equipment that works like this.

      • I figured sonar and the like was high powered and non-passive. But after seeing the link below from anonymous in another thread of the discussion I see the idea of passive acoustic sensing has been around long enough (pre WWII) to be highly developed.

    • by Anonymous Coward

      If "better acuity" is a euphemism for it won't shit on your lawn, then yes.

    • Re:Interesting... (Score:5, Insightful)

      by BitZtream ( 692029 ) on Thursday May 02, 2013 @10:37PM (#43616919)

      It is much more developed, you just don't realize it. This sort of signal processing is used in ultrasound machines, your car, my boat, fetal heartbeat monitors, little credit card readers that attach to your phone via headphone jacks. ALL SORTS OF STUFF.

      Its all just an application of some FFTs and some weighting. It isn't even non-trivial for anyone with some basic understandings.

      Without any prior knowledge, I went from no understanding to of FFTs to writing code to detect lean fuel during runtime on my R/C airplane. A Raspberry PI is also way more CPU power than needed to do it, which means it can also do a whole bunch of other stuff too.

      • by Black Parrot ( 19622 ) on Thursday May 02, 2013 @11:32PM (#43617187)

        A Raspberry PI is also way more CPU power than needed to do it, which means it can also do a whole bunch of other stuff too.

        My ++ model will mine Bitcoins between drone attacks!

      • Without any prior knowledge, I went from no understanding to of FFTs to writing code to detect lean fuel during runtime on my R/C airplane.

        That is fascinating. I feel motivated to go out and try myself, now

      • The ars article is light on details, but it looks like the author is leaning towards training a neural net to do the pattern matching since he mentions http://www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA320924 [dtic.mil]

        I probably wouldn't go the FFT route since a neural net or wavelets probably work better...but I also haven't done anything of the sort in the past 5-10 years, so maybe things got better. You're looking at the audio signature to determine lean fuel?

        • by rtb61 ( 674572 )

          Spend a bit more and go the simpler route of radio detection finding equipment. Drones are really quite radio chatty and sourcing a radio emission from altitude (video feed so the remote operator can see where they are going) tend to give them away.

    • Trained dog? Probably not. I guess it's possible, but probably not. But - you can put this piece of pie on a pole, or whatever, and forget about it. Toss it on your roof, or put it on your bird feeder, whatever. No need to walk it, water it, or much anything else. No fleas. It won't steal your covers in the middle of the night. No muddy tracks in the hallway. All that it will ever need is a supply of electrons!

      • Hehe, in some countries dogs are the low tech cheap option when theres tons of strays already. You'd be surprised at how easy some are to train, I would imagine some train themselves. We have a dog, and there are certain things that set her running and barking and its not the neighbors, its something halfway across the neighborhood. No way for us humans to tell what it was.

  • Range (Score:5, Insightful)

    by Dan East ( 318230 ) on Thursday May 02, 2013 @10:01PM (#43616697) Journal

    A Predator has an operating ceiling of 25,000 feet. You think a raspberry pi and mic is going to hear a Predator drone in cruise mode that's 5 miles above? You can't even hear a massive passenger jet at that altitude! Now a quadcopter is a different story, as they are as loud as can be, but saying this system would work on something like a Predator is a stretch.

    • Re:Range (Score:5, Interesting)

      by Anonymous Coward on Thursday May 02, 2013 @10:15PM (#43616785)

      Yeah, you'll need some bigger tools [greg.org] for that.

      • "Yeah, you'll need some bigger tools for that."

        Can't we just raise money to buy/rent/bribe the "bigger tools" who keep telling us that this (drone-use) horse$hit is "legal?" Just sayin'...

        • by KGIII ( 973947 ) <uninvolved@outlook.com> on Friday May 03, 2013 @05:05AM (#43618235) Journal

          I don't think that I disagree with your ideals. Having said that, and - really, I don't disagree with them in all likelihood, I must ask, "In which way(s) are you asserting that drones, or their use(s), illegal to start with? What, specifically, is illegal about drones? What laws are being broken with them?"

          I ask because, well, I fear that they will become so inexpensive to operate (including training) that they will be used to increase surveillance which, while against my ideals, probably doesn't (in and of itself) violate any laws. I'd like to be able to make a clear, factual, and reasonable argument against the increased use of drones but I don't have any arguments to make against it that are logical and truthful. I can only state that I'm afraid of what it may turn into.

          Even in my worst imagined fears concerning these drones, I'm unable to find any laws that are being broken. Perhaps you come from a different country than I do? I'm from the United States of America. We're not all that free here which means that there are a lot of laws. Given the number of laws it is impossible for me (I'm not a lawyer, even then it would still be impossible) to know all of the laws. However, I'm reasonably aware of the many laws that we have to protect us and I'm reasonably aware of the laws that the government must follow and I'm unable to point to anything specifically about drones (or their proposed use, or even with the uses that I'm afraid of) which are illegal.

          So, I'm not asking you to be my research assistant or the likes. You stated that drone use was illegal which isn't very specific and certainly isn't true so I'm hoping that you have something more concrete and that you can actually point out which laws being broken by these. If you could provide specifics and case law that would be most excellent too but I'm trying to not ask for too much.

          See, in my opinion, they are making their surveillance too intrusive and too easily accomplished. In this day and age we're able to be monitored in most everything we do and, while this has always been legal, it has become increasingly easy for this to be done and for this information to be shared. The use of drones by police, municipalities, and private companies to monitor, photograph, and track a person is just yet another step in the age of zero privacy. I'm quite certain that it (that portion specifically) is legal though I'm equally certain that it shouldn't be. I can vocalize, express, this as a worry and all and that may affect the opinions of some but if it is illegal then there's an even greater chance of demonstrating the harm that this can cause to society.

          I don't know... Perhaps you meant it should be illegal? I'm really unable to find anything that indicates the use of drones would be illegal in and of itself and that includes FAA regulations and the likes - I've looked. I could be missing something. Then again, this is /. and you could just be an insane zealot who thinks that anything they don't like is illegal but I'm hoping you're more honest and logical than that. Trust me on this, I've seen it here before...

          Either way, I'm hoping you have something more concrete than the generic statement that the use of drones is illegal because, well... No, no it isn't illegal in and of itself. If there's some specific manner that they're being used that is illegal then I'm quite interested in knowing what it is. Like I said, I've even met people online who thought that DST (Daylight Savings Time) was illegal so you could just be one of the crazy ones but I'm sort of hoping you're not. It would be, admittedly, amusing but it wouldn't actually be beneficial to me.

    • Re:Range (Score:4, Insightful)

      by multiben ( 1916126 ) on Thursday May 02, 2013 @10:15PM (#43616789)
      Just because you can't hear it doesn't mean a microphone can't pick it up. I am not going to google this because I am lazy, but I would not be surprised at all if a good quality, well aimed, highly directional mic could pick up enough of a noise to run an analysis.
      • Re:Range (Score:5, Funny)

        by Anonymous Coward on Thursday May 02, 2013 @10:23PM (#43616841)

        So now the highly directional microphone has to be pointed toward the undetected drone in order to detect it? That makes perfect sense.

        • Hehe, yeah, but a whole bunch of these all over the place in a distributed network that all dumped to one database could make a very interesting historical map of drone flight paths.

          • Re:Range (Score:5, Funny)

            by Cryacin ( 657549 ) on Friday May 03, 2013 @12:05AM (#43617335)
            Dusts off old chestnut

            Sounds like you need a beowulf cluster for that!
          • by AHuxley ( 892839 )
            Yes flay, You could build a huge network over a wide area, with wired connections to each mic. Back to a central hub for very basic noise values math.
            The best location would be in a more empty state with lots of new drone 'schools' for maintenance, new staff needing many flight hours. This would give you some open space, a wide sky to scan and that real sound over many months.
            Due to the nature of drone training, one would expect this to be away from suburbia, the big air traffic hubs. Based on video cap
            • Aye I was thinking the idea would be for a more opt in approach and people would do either a seti at home type application or it would get sent off to a nerd somewhere with some computing power. Since the gentleman running the project picked the raspberry pi as a platform it already has cellular capability to phone home collected data over time.

              Though personally I would be a bit iffy about leaving my rasberry pi out on a mountain range somewhere un-attended. Also the places you mention that would be good sp

        • Re:Range (Score:5, Interesting)

          by fluffy99 ( 870997 ) on Thursday May 02, 2013 @11:11PM (#43617097)

          So now the highly directional microphone has to be pointed toward the undetected drone in order to detect it? That makes perfect sense.

          If you're sophisticated and have multiple microphones in a well planned array, then you can aim your microphone in software and sweep the sky looking for the signature. Look up acoustic beam-forming. If the array is large enough you can estimate distance as well as angle. The bonus is that you get actual tracking instead of just detection.

          The problem would be processing power though. Simple implementations could range from 4 microphones that you sum/subtract to look at quandrants, up the way to something approaching what the US Navy does with its towed arrays. I doubt the PI could handle the processing of the signals in both the time domain to get tracking, and the frequency domain to do target qualification.

          Of course you also have the question of what do you do when you detect one? Aim a camera at it? Fire off your green laser? (no not suggesting you commit a felony).

          • Re:Range (Score:5, Informative)

            by DerekLyons ( 302214 ) <fairwater@@@gmail...com> on Friday May 03, 2013 @02:42AM (#43617799) Homepage

            Disclaimer, I was a submarine crewman and though I was a qualified sonar watchstander I was not a sonar tech. I have also studied fairly widely in the unclassified literature.

            So now the highly directional microphone has to be pointed toward the undetected drone in order to detect it? That makes perfect sense.

            If you're sophisticated and have multiple microphones in a well planned array, then you can aim your microphone in software and sweep the sky looking for the signature. Look up acoustic beam-forming. If the array is large enough you can estimate distance as well as angle. The bonus is that you get actual tracking instead of just detection.

            It's certainly simple - in theory. In reality, picking out such a faint noise from the background is Very Difficult Indeed.
             

            The problem would be processing power though. Simple implementations could range from 4 microphones that you sum/subtract to look at quandrants, up the way to something approaching what the US Navy does with its towed arrays. I doubt the PI could handle the processing of the signals in both the time domain to get tracking, and the frequency domain to do target qualification.

            Processing power, both for signal analysis (finding the faint signal) and for beamforming is on the beginning of your problems. Let's just hit the high spots:

            • The accuracy of your track is only as good as the accuracy of your microphone positioning. (You won't need surveyor grade accuracy, but you will probably need better than the three meter accuracy that WAAS/GPS provides.) You can't beamform if you don't know the relative locations of your microphones. Oh, and did I mention that sound is refracted as the temperature of the air changes? You'll have to account for that too - assuming you can get accurate enough data on current conditions.
            • You'll need some fairly clever filtering and processing to avoid the microphones being swamped by unrelated and louder background noise.
            • You also need high quality low noise amplifiers to bring the sound of Predator up to useable levels. (The highest quality commercial audiophile amplifier isn't even close. You need a supercomputer and audiophile gear by comparison isn't even as good as the throwaway calculators you get with your breakfast cereal.)
            • Speaking of the sound... different frequencies get attenuated and refracted differently. You'll have to account for that too.

            Etc... etc... The very definition of a non trivial project. You're essentially trying to replicate what the USN does with it's passive sonar systems, with dull and chipped stone knives. (You don't even have a bearskin. You don't even get a bearskin, just the aforementioned knives.)

            • by chihowa ( 366380 )

              The accuracy of your track is only as good as the accuracy of your microphone positioning. (You won't need surveyor grade accuracy, but you will probably need better than the three meter accuracy that WAAS/GPS provides.) You can't beamform if you don't know the relative locations of your microphones. Oh, and did I mention that sound is refracted as the temperature of the air changes? You'll have to account for that too - assuming you can get accurate enough data on current conditions.

              I like solving problems and this is somewhat similar to what I do for a living, so I'll speculate...

              In a similar approach to the use of a guide star [wikipedia.org] in astronomy, you could use an airliner flying overhead (or a helicopter for a coarse calibration) to calibrate your microphone array and correct for changes in refraction. An airliner or helicopter will be easy to see with a camera, and of a known size, altitude, and speed.

          • by xelah ( 176252 )
            I think that if you did do that, at least in certain countries, you'd discover it to be a very good way of making yourself the target of the next extrajudicial killing by the US state. Unless you think you could do all that without it being obvious to surveillance or any nearby intelligence services?
            • This is why you automate it. So you can blind the drone's camera, but be far enough away from the site when the missiles hit. With the right cammo, the detector is only visible when it fires the laser, which it doesn't do unless it hears a drone. Passive surveillance/active response system for asymmetrical warfare.

        • "So now the highly directional microphone has to be pointed toward the undetected drone in order to detect it? That makes perfect sense."

          Ever seen the rotating antenna of the radar on a boat? That's the reason it rotates.
          To detect undetected stuff.

          I guess if many people are ordering one of these drone detectors from a cave in Pakistan, it could be a clue that it works.

    • A quadcopter at $25k feet? Impressive, that I would like to see.

      You can't hear loud noises in the same room with you if you've tuned them out, just because your mind makes it out to be irrelevant background noise doesn't mean that its not there.

      The sound can be orders of magnitude lower in amplitude than the noise floor and DSPs can still pick up the signal.

      Your cell phone works with far worse signal to noise ratios than a mic finding a drone at 25k vertical.

    • by Rich0 ( 548339 )

      I'd think that a distributed passive RF detection system would be more useful.

      Just about any aircraft reflects radio waves (even stealth aircraft - they just reflect them away from the transmitter). In theory passive devices could capture RF like a radar and look for reflections, and a bunch of these in different places could determine the position of any aircraft that reflects RF. That would likely have a range much larger than sound.

  • by Anonymous Coward

    You run away?

    That's as much of a "shield" as a radar detector "protects" you from speeding tickets.

    • You run away?

      That's as much of a "shield" as a radar detector "protects" you from speeding tickets.

      OK, have it send the e-mail to your next of kin.

    • Re:And then what? (Score:4, Interesting)

      by AK Marc ( 707885 ) on Thursday May 02, 2013 @11:55PM (#43617299)
      The greatest use of drones is still reconnaissance. So you stop doing the illegal things until it's gone. Why run from a camera? Just hide your activities until it's gone.
      • by Rich0 ( 548339 )

        Well, I suspect that in a few years the thing will basically tell you that there is a drone overhead 24x7. So, those with tinfoil hats basically will end up never going outside.

    • by TedRiot ( 899157 )
      Fire your (directional) EMP generator at the drone, of course.
    • A radar detector can protect you from speeding tickets. A laser detector...not so much.

  • Tough (Score:5, Insightful)

    by Cassini2 ( 956052 ) on Thursday May 02, 2013 @10:03PM (#43616705)

    It's tough to reliably detect low-level background repetitive noise without detecting all sorts of nearby domestic appliances, car engines, and such. In the modern city, we live with *alot* of noise.

    Now, if the problem is to detect jet engines in rural areas featuring mountainous terrain, then I think I know what the point of this project is.

    • It wouldn't be that bad.

      Have a 360Â servo with a 80Â-90Â servo mounted on it. Directional mic on top of all that.

      Arduino/Pi rotates the first servo 1Â then sweeps the second servo. Or vice versa.

      Feed that into an algorithm looking for prop noise. Most drone motors will be IC or electric. An IC will be running between 9K~18K RPM. Electric would be running from 7K on the low end to 30K on the high end. Realistically, an electric for drone use would be on the low end of that spectrum;

  • by Anonymous Coward

    Someone comes along playing the sound of a drone on their smartphone and you realized what a waste of time and money it was.

    • Sure, and they'll be able to hold that smartphone up in front of another phone and play modem sounds and they'll be able to communicate at full 56k speeds!

      Not really, in both cases.

      You can't tell the difference, doesn't mean that the signal processor can't.

    • Someone comes along playing the sound of a drone on their smartphone and you realized what a waste of time and money it was.

      Here you go! [bbc.co.uk] Down near "Sound can be a "two-sided coin" in war, he argues."

  • by porkThreeWays ( 895269 ) on Thursday May 02, 2013 @10:10PM (#43616755)
    So, it's a project with a computer that happens to run on a slightly slower processor. If this ran on a mini-itx no one would even mention that part.
    • by MacTO ( 1161105 )

      Of course, the Pi is also smaller and uses less power than a mini-itx board. Sticking something the size of a small tissue box in the backyard, and doesn't have an AC cord running into the house, is going to create a lot less friction with the spouse. Many people also find solving problems with constrained resources fun.

    • At least it's not an arduino.

  • by girlinatrainingbra ( 2738457 ) on Thursday May 02, 2013 @10:17PM (#43616797)
    This is more likely to work better as a "squirrel" vs. "bird" detector, or with good/better datasets, perhaps even as a "cardinal" vs. "bluejay" vs. "mockingbird" vs. "car alarm" detector, especially if the birds are in your front or back yard. But really, the concept of hearing a predator drone is very farfetched, unless the drone is flying super low for some reason!
    .
    Better to use this as an auto-logging device for some birders falling in love with counting how many birdies are coming by, or for recording to the exact micro-second when the swallows [wikipedia.org] finally make it back to San Juan Capistrano [wikipedia.org]!
    • by ceoyoyo ( 59147 ) on Thursday May 02, 2013 @10:54PM (#43617027)

      Hm. That might be interesting. Build a couple and put them a distance apart, plus a remote controlled super soaker. Sit in the basement and pretend it's a sub.

      "Skipper, we have a contact on bearing 238... probable squirrel class mammal, likely a grey!"

      "Do you have a firing solution?"

      "Range and course calculated skipper! Firing solution locked in!"

      "Fire!"

      "Skipper, sudden change of aspect on the contact. I have angry squirrel sounds."

      "Nice job, everybody."

    • This is more likely to work better as a "squirrel" vs. "bird" detector, or with good/better datasets, perhaps even as a "cardinal" vs. "bluejay" vs. "mockingbird" vs. "car alarm" detector, especially if the birds are in your front or back yard.

      That is a much more practical and marketable idea. Now go build it!

    • Hmmm...how about something www.capistranoswallows.com running on a raspberry pi to present a webpage like http://www.abevigoda.com/ [abevigoda.com]

  • Don't drones blast RF on known frequencies? Unless they're flying with auto pilot and just collecting data, but that would be dangerous in an urban environment with lot's of air traffic. How much would it cost to build something with an antenna that's just listening for broadcasts on these frequencies? You could even measure the doppler shift to see if they're coming or going, and at what speed. Listening for audible noise just seems useless anywhere outside of a big empty desert with a large directional mi

    • There are many types of drones. You're thinking of a more radio controlled type. The better ones just have a cellular chip in them and you control them over the internet. Rather than "Fly them" like an RC plane, you give it a target and it goes on its merry way. Once it has its instruction set it doesn't even need the cellular connection anymore, it can just fly back to "home base" once it has done whatever it is it was supposed to do.

    • I think they are usually blasting up using sat uplink (military drones). Civilian ones use ordinary RC radios - good luck distinguishing between cordless phone, bluetooth and Fatuba working on 2.4GHz.

    • by AHuxley ( 892839 )
      Depends on the drone.
      Some talk to a van direct. Great fun if the bad guys ever triangulate the base team :)
      Some long lasting loitering surveillance aircraft with a look down link gets more interesting. But that might be noticed long term.
      Other option is a satellite dish like link. Drone only transmits 'up', the van is half a world away.
      Long term expect a cheap blimp to sit over an area of interest. Just like a satellite dish link but massive bandwidth and very easy to add more blimps :)
      Almost aut
      • I imagine really smart reconnaissance drones would fly their mission then return to a safe zone behind your line to transmit and report home. Real time reconnaissance in an actual combat zone would be supported by the more directly controlled drones in which case finding the controller might be nice, but futile as well.

    • Predators and Reapers at least do transmit a satellite video feed which can be picked up from the ground (completely unencrypted, last I heard). So yeah that's one "radio warning sign" that could be worth looking out for. A computer with a TV-tuner-based SDR could detect it. Even if they encrypt the feed, it would be worth knowing about any new satellite video signals that suddenly show up.

  • Seriously, when I read the headline, my mind read "Meat Drone Shield" and I thought at first there was some kind of organic meat shield which screwed up with drone targeting systems (like infra-red, etc.) and provided protection.
  • Comment removed based on user account deletion
  • Isn't the government supposed to be in service to the citizens? And here are the citizens getting organised and generating their own funding in order to defend themselves from the government. That's the same government which is 'elected' by the citizens. You know things are outrageously out of hand when citizens are funding the generation of defensive mechanisms to protect themselves from the military industrial complex of their government.

    Isn't it about time for honest Americans to stand up and say "I d
  • better idea (Score:4, Funny)

    by slashmydots ( 2189826 ) on Friday May 03, 2013 @12:22AM (#43617391)
    You actually don't need to listen for them. If you watch really carefully, you might be able to see part of your neighborhood blow up. Then you know a drone is overhead.
  • "Warning: Predator drone detec---"

  • Predator drones use a common small aircraft engine used on about 30 other small aircraft. Similar Rotax engines are used in snowmobiles, jet skis, go-carts, and other small engine applications.

    You're going to get an alarm every time something with a small engine goes by, and you probably won't pick up an aircraft flying high enough to not be blatantly annoying.

    If you want to detect nearby aircraft, build a radar. There are automotive anti-collision radars that could be adapted.

    • by Genda ( 560240 )

      It would be fun to build a small radar array with a synthetic aperture antenna so you could tell the difference from let's say a Cessna 152 flying by and a drone. The next step would be to use the spacial information to aim a small high performance maser to fry the electronics on said drone (of course, only if it was invading your privacy.) It would be entertaining for the drones' owners to find in post mortem that their sky spy was a crispy critter.

    • There aren't many snowmobiles, jet skis, or go-karts operating in the areas where drones are operating. Especially in the air.

      At the same time, I don't believe that audible signals would be effective since a drone operates several miles above the ground surface.

      What is needed for drone detection and identification is a wide-angle magnified video camera that scans the sky (basically the same thing as the Predator uses on its targets). A Predator is about 27x48 feet (length x wingspan), at 20x magnification

      • There aren't many snowmobiles, jet skis, or go-karts operating in the areas where drones are operating. Especially in the air. I got a great chuckle out of this one =)

        The camera's needed would maybe be a lot more expensive then your standard 3-6 megapixel cheap ones. Once you start getting into good resolutions and light adaptability. It starts getting harder. Then again maybe an array of off the shelf webcams could do the trick, but that requires group participation and you think people are mad now about g

  • Apart from hobby drones (like quadcopters) it's fairly limited what drones you'll encounter around where you live. Unless you live where the wild Predator drones roam... there you'll most likely hear the sound of missiles detonating before you hear the drone itself.

  • Great, because I was worried about these. [xkcd.com]
  • Given the Predator uses hellfire missiles that are supersonic, they're going to blow you up before it is physically possible to hear them. And the drone itself can fire from five miles away.

I have hardly ever known a mathematician who was capable of reasoning. -- Plato

Working...