Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Your Phone Can Be Snooped On Using Its Gyroscope

Unknown Lamer posted about a month ago | from the phone-can-be-snooped-on-by-everything dept.

Cellphones 96

stephendavion (2872091) writes Researchers will demonstrate the process used to spy on smartphones using gyroscopes at Usenix Security event on August 22, 2014. Researchers from Stanford and a defense research group at Rafael will demonstrate a way to spy on smartphones using gyroscopes at Usenix Security event on August 22, 2014. According to the "Gyrophone: Recognizing Speech From Gyroscope Signals" study, the gyroscopes integrated into smartphones were sensitive enough to enable some sound waves to be picked up, transforming them into crude microphones.

cancel ×

96 comments

Sorry! There are no comments related to the filter you selected.

Is it "Gyro" or "Yeero"? (1)

Anonymous Coward | about a month ago | (#47711415)

Is it "Gyro" or "Yeero"?

Re:Is it "Gyro" or "Yeero"? (0)

Guspaz (556486) | about a month ago | (#47711613)

It's "Tasty".

Re:Is it "Gyro" or "Yeero"? (1)

Pope (17780) | about a month ago | (#47713847)

So, you like-a the sauce?

I give up... (4, Funny)

fuzzyfuzzyfungus (1223518) | about a month ago | (#47711423)

Can we just succumb to the inevitable and work on building a list of the parts of a smartphone that can't be used to spy on you?

I'm thinking 'maybe the battery door'. Any other suggestions?

Re:I give up... (1)

Anonymous Coward | about a month ago | (#47711475)

They're probably working on the battery door spying technology as we speak.

Re:I give up... (3, Funny)

nxtr (813179) | about a month ago | (#47711593)

The vibrations of your voice cause voltage fluctuations in the battery, and can be reconstruct the image of the suspect from the reflection in the eye of the person across the street.

Re:I give up... (1)

fuzzyfuzzyfungus (1223518) | about a month ago | (#47711989)

Given that attack [tau.ac.il] where they used the acoustic emissions of a CPU's voltage regulator circuitry to extract an RSA key I'm going to consider that one guilty until proven innocent...

Re:I give up... (0)

Anonymous Coward | about a month ago | (#47714533)

Einstein's law of conversation of energy working at full capacity here.

Any vibration can be used as a mic. It's the physics of the situation--the problem is the noise that typically comes in a signal.

Re:I give up... (3, Funny)

plover (150551) | about a month ago | (#47711503)

Apple fixed that problem. The iPhone has no battery door, so it can't be removed in case you don't want the phone to spy on you.

Re:I give up... (5, Insightful)

geekmux (1040042) | about a month ago | (#47711667)

Can we just succumb to the inevitable and work on building a list of the parts of a smartphone that can't be used to spy on you? I'm thinking 'maybe the battery door'. Any other suggestions?

What's the point of securing any smartphone when all of your activity on the device is captured elsewhere and sold for profit? They don't just count how many times you play your songs. They count how many times you text during the day. They count how many times you click on icons. They count how many seconds you hover over app icons even when you don't buy them in order to market apps catered to your "maybe" whims. Yes, they do this shit. No, it's not called crazy, it's called statistical analysis to the nth degree in order to maximize profits.

The phone is merely the vehicle. What that vehicle can do all depends on the driver. Unfortunately, we've all been thrown in the back of a telco cab and the driver was told to get lost years ago.

A profitable business idea ... (0)

Anonymous Coward | about a month ago | (#47712931)

... maybe even eligible for a genuine patent (not that I favor patent, but ...)

Can we just succumb to the inevitable and work on building a list of the parts of a smartphone that can't be used to spy on you? I'm thinking 'maybe the battery door'. Any other suggestions?

What's the point of securing any smartphone when all of your activity on the device is captured elsewhere and sold for profit?

The point being there IS an opportunity for anyone who comes up with a workable idea to really really lock down all your gadgets (not only smartphones but all electronic gadgets) so that even when the gadgets are powered up they can't leak _any_ information

Re: A profitable business idea ... (2)

lemonfresh33 (1367367) | about a month ago | (#47712953)

Just don't use that device anywhere public. Or on a public network because they can snoop on you that way.

Re: A profitable business idea ... (1)

geekmux (1040042) | about a month ago | (#47713345)

Just don't use that device anywhere public. Or on a public network because they can snoop on you that way.

Yes, I'll just go home and surf. I'm sure I'll be perfectly safe from spying there.

After all, I trust my ISP so much that I don't even consider them a "public" network anymore. They gave me a custom home page that goes right to THEIR website, so it must be private, right? And look here, this systray icon even has their logo! I am so loved I'm practically an employee.

Re: A profitable business idea ... (1)

RabidReindeer (2625839) | about a month ago | (#47713583)

Just don't use that device anywhere public. Or on a public network because they can snoop on you that way.

Yes, I'll just go home and surf. I'm sure I'll be perfectly safe from spying there.

After all, I trust my ISP so much that I don't even consider them a "public" network anymore. They gave me a custom home page that goes right to THEIR website, so it must be private, right? And look here, this systray icon even has their logo! I am so loved I'm practically an employee.

Mine too! https://room614a.att.com/ [att.com] . It's SSL, so I know I can trust it!

Re:A profitable business idea ... (1)

geekmux (1040042) | about a month ago | (#47713139)

... maybe even eligible for a genuine patent (not that I favor patent, but ...)

Can we just succumb to the inevitable and work on building a list of the parts of a smartphone that can't be used to spy on you? I'm thinking 'maybe the battery door'. Any other suggestions?

What's the point of securing any smartphone when all of your activity on the device is captured elsewhere and sold for profit?

The point being there IS an opportunity for anyone who comes up with a workable idea to really really lock down all your gadgets (not only smartphones but all electronic gadgets) so that even when the gadgets are powered up they can't leak _any_ information

Yes, there is. And there are companies that are attempting to offer secure services and devices like this, such as Silent Circle and Blackphone.

However, your mistake with this "profitable" business idea is thinking that the majority of people actually give a shit about security and privacy and will PAY for such a service.

The current environment was birthed from the ignorance that they don't. And won't. Only a small fraction of people care enough to pay, which may or may not be profitable enough to even attempt.

On top of that, vendors are practically drooling over the onslaught of IoT devices coming, and soon it will literally be illegal for you to make a "silencer" for an electronic device. The manufacturers will lobby to ensure this happens. And they'll win.

Re:I give up... (1)

antdude (79039) | about a month ago | (#47716231)

I am so glad I don't own a mobile phone to worry about these issues. :) I do have to worry about other sources though. :(

Re:I give up... (0)

Anonymous Coward | about a month ago | (#47712243)

This is why RMS doesn't like any type of cell phone. He considers it Stalin-esque to require people to carry a radio transmitter anywhere they go. But everybody does it voluntarily, with a microphone attached no less! Frankly, if a state actor wants to spy on you, they will, full stop. There are even ways of bugging typewriters by having a microphone in the room. The acoustic pattern for each key is different. The way to be secure with a smartphone is to chuck it into the river and never replace it, then keep your head down for the rest of your life.

Re:I give up... (1)

viperidaenz (2515578) | about a month ago | (#47714743)

But my cellphone has two microphones!

It's also waterproof, so what's a river going to do to it?

Re: I give up... (1)

jd2112 (1535857) | about a month ago | (#47712253)

So That must be the reason that Apple made the iPhone battery non removable. It's a security feature!

not the battery door (1)

oneiros27 (46144) | about a month ago | (#47712469)

Mine's got a wireless charging pad in it.

Of course, it's running WebOS, which lets me set up security such that I can require confirmation before an app's allowed to use certain features (eg, GPS), rather than just giving it a blanket 'you're allowed to use GPS whenever you want to'.

The drawback is that I don't have nearly as many apps available to use, being that it's WebOS. (I still blame those horrible Palm Pre commercials with the stoned albino -- why they didn't bother showing that it supported multitasking and copy & paste way before iOS, I have no clue)

Re:not the battery door (1)

viperidaenz (2515578) | about a month ago | (#47714767)

A wireless charging pad? So they can just listen for the power consumption of the CPU with an RF antenna, process the waveforms to extract encryption keys and then hack in via the cellular radio and take information they want?

Re:not the battery door (1)

fuzzyfuzzyfungus (1223518) | about a month ago | (#47727657)

As much as I mourn my HP Touchpad (Oh man did WebOS multitasking curb-stomp Android multitasking at the time and even considerably later); if you are still running WebOS you probably have bigger security issues. The last update for any Pre models was December 2011, and Touchpad models January 2012. That's a long time for a relatively full featured OS to go without any fixing.

Re:I give up... (1)

schlachter (862210) | about a month ago | (#47712545)

it will be more dangerous when they figure out how to make your battery explode or electrocute you. when the go from surveillance to attack mode.

Re:I give up... (1)

camperdave (969942) | about a month ago | (#47712645)

It won't work. They'll be able to track you through the "hole" you leave by not having a trackable smart phone.

Re:I give up... (1)

thieh (3654731) | about a month ago | (#47713727)

We probably should stop using cellphones altogether.

Re:I give up... (1)

Hattmannen (658936) | about a month ago | (#47713933)

Battery door? Oh, that thing on the back of the phone. Isn't that that elusive back door I've been hearing so much about? :-p

Cutting edge news? (0, Insightful)

Anonymous Coward | about a month ago | (#47711453)

That's modern /. for you. They should change the motto to "Week Old News for Nerds".

Re:Cutting edge news? (0)

oodaloop (1229816) | about a month ago | (#47711487)

The event isn't until 2 days from now. That's ludicrous speed for /..

Re:Cutting edge news? (0)

Anonymous Coward | about a month ago | (#47711639)

One must travel at ludicrous speed to get news for nerds from the future. I'm guessing it was gyroscopes themselves spinning so fast as to open a wormhole to Dice and play the Usenix Security session to Unknown Lamer.

I've only managed to do this once myself.

Re:Cutting edge news? (0)

Anonymous Coward | about a month ago | (#47711509)

Or perhaps more apropos: "Weak News for Nerds"

Re:Cutting edge news? (1)

Anonymous Coward | about a month ago | (#47711527)

You are so much smarter for having found this information a week ago. It's practically useless now.

So? (0)

rhazz (2853871) | about a month ago | (#47711491)

I'm going to assume most phones already have actual microphones, so how does this add any additional kind of insecurity? I'm going to assume most phones already have actual microphones, so how does this add any additional kind of insecurity?

Re:So? (2)

plover (150551) | about a month ago | (#47711519)

I'm going to assume most phones already have actual microphones, so how does this add any additional kind of insecurity? I'm going to assume most phones already have actual microphones, so how does this add any additional kind of insecurity?

Apparently the sound from your mic and the echo from your gyroscopes were both parsed by your speech-to-text converter. I guess it works better than we thought!

Re:So? (0)

Anonymous Coward | about a month ago | (#47712299)

Woooosh!

Re:So? (3, Informative)

JazzXP (770338) | about a month ago | (#47711535)

Basically an app can ask for permissions for the gyro only (if it even needs to) and be recording conversation.

Re:So? (3, Interesting)

rasmusbr (2186518) | about a month ago | (#47711843)

Basically an app can ask for permissions for the gyro only (if it even needs to) and be recording conversation.

Yeah, that's the thing. You don't need permissions for the gyro on Android and iOS, so any and all of the apps that you have on your phone or tablet could be using the gyro and you wouldn't know, except for an anomalous battery drain.

Re:So? (3, Interesting)

frinkster (149158) | about a month ago | (#47712115)

Basically an app can ask for permissions for the gyro only (if it even needs to) and be recording conversation.

Yeah, that's the thing. You don't need permissions for the gyro on Android and iOS, so any and all of the apps that you have on your phone or tablet could be using the gyro and you wouldn't know, except for an anomalous battery drain.

Sure, but on iOS an app is suspended when you are on a phone call unless the app has used the system APIs to enable background execution. There are only a small number of background execution modes and your app must declare which it plans to use. When it comes to location-based background execution (the most likely use of the gyro), your app still gets suspended. The system wakes it up periodically and sends location updates to a function in your app and then gives the app a small time window for that function to return an expected value. It is very much a discrete task-based multitasking system - completely different than normal desktop machines. Good sometimes. Bad sometimes.

Re:So? (2)

rasmusbr (2186518) | about a month ago | (#47712631)

Permissions on Android are a bit more rudimentary, so it would be simple to make a background process that just sits and quietly listens to the gyro. You would need to ask for the permission to keep the device awake in order to keep the CPU and sensor chip alive and (in order for it to be practical) the permission to start on boot.

Re:So? (1)

RabidReindeer (2625839) | about a month ago | (#47713719)

The point of all this isn't to record phone conversations. Some of the agencies likely to exploit this particular weakness have more than enough clout to tap the main communications channel for that.

The accelerometer exploit is a very low-quality audio sampler. Sample range tops out at about 200/second, IIRC. Enough to get a muffled audio, but nowhere near opera-quality.

However, it's something that someone could do to monitor room conversations when the phone isn't on a call. And current access controls don't provide enough protection. One of the most promising solutions is simply to limit sample rates to something relatively useless like 50/second.

Then again, if all you want is to detect basic acoustical vibrations, even that would be enough to tell people to turn on their "houseplant audio monitors".

You can also get good audio off the reflections of a tinfoil hat.

Re:So? (1)

Anonymous Coward | about a month ago | (#47711549)

Some of my co-workers were talking about this last week, and I think the effective issue was that while accessing the microphone requires special privileges to be granted to the application, no such privileges are required to access the gyroscope.

Re:So? (0)

Anonymous Coward | about a month ago | (#47711589)

an app generally can't record your phone calls and covertly send that data back... just about every app in the world has access to the gyroscope (for games!), so T.H.E.Y. (or T.H.E.M.) could be listening in on your phone calls.

Re:So? (0)

Anonymous Coward | about a month ago | (#47711673)

IF you RTFA, you'd know that OS restricts app access to microphones. No such restriction, so far, exists for the gyroscope.

Re:So? (1)

rhazz (2853871) | about a month ago | (#47719687)

Guilty as charged.

what is gyrating inside mobile devices? (0)

Anonymous Coward | about a month ago | (#47711499)

It feels somewhat cumbersome to read about groscopes in mobile devices. Ah well, there is also no disks inside SSD hard disks...

Oh no (1)

Parker Lewis (999165) | about a month ago | (#47711511)

Another summary posted by trainee.

Hard to be sure (0)

Anonymous Coward | about a month ago | (#47711513)

But from what I gather, Researchers will demonstrate a way to spy on smartphones using gyroscopes at Usenix Security event on August 22, 2014.

Phones have microphones (0)

Anonymous Coward | about a month ago | (#47711517)

I'm not sure why anyone would care that some aspect of their device could be used as a "crude microphone," when a perfectly functional and often well-designed microphone is an integral part of the design. Seems pretty silly.

App permissions (2)

grimJester (890090) | about a month ago | (#47711809)

Apps request permissions for different pieces of hardware on a case-by-case basis. The average user might raise some eyebrows if an app that shouldn't need it wants to access your microphone, but access to gyroscope data might not even require user acceptance.

Re:App permissions (0)

Anonymous Coward | about a month ago | (#47712215)

Neither iOS nor Android require user permission to access the gyroscope data. It's not even in the prompts.

Hiding the phone... (3, Funny)

khr (708262) | about a month ago | (#47711533)

the gyroscopes integrated into smartphones were sensitive enough to enable some sound waves to be picked up, transforming them into crude microphones

Yeah, that's why I always stick my phone inside an empty potato chip bag when I'm talking to someone...

Re:Hiding the phone... (0)

Anonymous Coward | about a month ago | (#47711595)

That's why you should stick with Apple. With an iPhone you can get the same effect simply by holding your phone norma^W^W^W^W^Wwrong.

Re:Hiding the phone... (1)

Dog-Cow (21281) | about a month ago | (#47712267)

Being an un-funny ass must be really hard work these days. If we were to apply the oh-so-humorous ^W's in your post, it would read That's why you should stick with Apple. With an iPhone you can get the same effect simply wrong. Which, as we can see, makes no sense at all.

Re:Hiding the phone... (0)

Anonymous Coward | about a month ago | (#47712439)

Oops. Well, you have to admit: I did get the effect simply wrong.

Whew! (1)

Somebody Is Using My (985418) | about a month ago | (#47711571)

My phone doesn't have gyroscope, therefore I am safe from people listening in to my conversations.

"Gee boss, we need to spy on this guy! Any ideas how we can do it?
"Well he has a smart phone; maybe we can leverage that to our advantage?"
"Oh, I see what you are getting at; we'll hack the firmware so we can use the oscillations of the GPS to crudely and inaccurately record what he is saying!"
"Actually, I was thinking we might want to use the attached microphone which is, you know, designed to pick up sound..."

How to summary (1)

skovnymfe (1671822) | about a month ago | (#47711581)

Researchers will demonstrate the process used to spy on smartphones using gyroscopes at Usenix Security event on August 22, 2014. Researchers from Stanford and a defense research group at Rafael will demonstrate a way to spy on smartphones using gyroscopes at Usenix Security event on August 22, 2014.

Why the redundancy? Post must be longer than 100 characters?

Re:How to summary (1)

jheath314 (916607) | about a month ago | (#47711803)

"is actually a quined phrase that makes its point via repetition" is actually a quined phrase that makes its point via repetition.

Re:How to summary (1)

ArcadeMan (2766669) | about a month ago | (#47711943)

The post was written by The Department Of Redundancy Department.

this has been a passive attack for a while now. (1)

nimbius (983462) | about a month ago | (#47711599)

originally researchers analyzed the data in the following categories. wobbling or wiggling in android devices indicated stress patterns, while violent shaking concluded frustration or rage. Finally, a single impact for iPhone devices registered as a trip to the genius bar and an unpaid credit card bill.

The paper says... (4, Informative)

Anonymous Coward | about a month ago | (#47711649)

They are currently able to recognize the spoken digits 1-9 correctly approximately 80% of the time. This is given a training data set from the same speaker and the same phone. Incredibly impressive, especially since it was done from a web browser and requires no special permissions or even knowledge from the user. For those of you that didn't read it. However, James Bond spy tool this is not yet...

Re:The paper says... (1)

F.Ultra (1673484) | about a month ago | (#47712295)

Something tells me that this could quite easily be fixed by filtering out "noise" from the gyroscopes before presenting it to apps. There can hardly be a use case for this finegrained details from the gyro except this one of course.

Re:The paper says... (0)

Anonymous Coward | about a month ago | (#47712749)

except this one of course

Or any other use case that needs raw gyro readings. These use cases are common enough that Apple added a high-performance low-latency API for the gyro and accelerometer data because their high-latency filtered version wasn't good enough. Google "Kalman filter". If you want to do sensor fusion on gyro and accelerometer you're going to get drift, and you can eliminate the drift with the camera, but it's too bleeding-edge (and expensive) to be an OS service, so the OS needs to offer the raw data to anyone wanting to implement their own Kalman filter for this purpose.

Re:The paper says... (1)

F.Ultra (1673484) | about a month ago | (#47724439)

And we really need "Kalman filters" on our phones? And there is no way in hell that one could filter out the noise level produced by audio (which must be extremely low) and still give enough resolution for the Kalman filter ?

The paper says... (0)

Anonymous Coward | about a month ago | (#47713163)

Oh good ... a proof of concept from researchers with limited resources is only reliable 80% of the time. Nobody would want to improve on that. Nothing to see here.

Re:The paper says... (1)

macklin01 (760841) | about a month ago | (#47713351)

It would be neat, however, to see gyroscope inputs added to regular audio inputs to improve speech-to-text. This seems to be a nice proof of concept for that.

Re:The paper says... (1)

Pope (17780) | about a month ago | (#47713867)

So the Numbers Station folks aren't out of a job yet!

Re:The paper says... (0)

Anonymous Coward | about a month ago | (#47716115)

Even with Firefox at 200 samples/sec you can't get anything over 100 Hz without aliasing. Since voice is above 80 Hz this is just barely possible to get very low quality voice for a few people, if they are speaking loudly and there is no other noise at all.

With Safari at 20 samples/sec there's no way to reconstruct voice from the web browser, unless you are listening to elephants.

But snooped on with what? (3, Interesting)

sacrilicious (316896) | about a month ago | (#47711743)

Researchers will demonstrate the process used to spy on smartphones using gyroscopes at Usenix Security event on August 22, 2014. Researchers from Stanford and a defense research group at Rafael will demonstrate a way to spy on smartphones using gyroscopes at Usenix Security event on August 22, 2014. According to the "Gyrophone: Recognizing Speech From Gyroscope Signals" study, the gyroscopes integrated into smartphones were sensitive enough to enable some sound waves to be picked up, transforming them into crude microphones.

I can't help but feel like there are gyroscopes involved in this process somehow...

Re:But snooped on with what? (1)

wonkey_monkey (2592601) | about a month ago | (#47712169)

Well, they call 'em gyroscopes, but since gyro- implies something spinning, and these things are (as far as I can ascertain) just vibrating, I call 'em accelerometers.

And real gyroscopes have more purposes than measuring acceleration anyway.

Re:But snooped on with what? (1)

oodaloop (1229816) | about a month ago | (#47712611)

The gyroscopes and accelerometers are two different things. The gryoscope measures tilt of the device, like when you play a racing game. The accelerometer measures change in velocity, like when you shake the phone to shuffle a playlist. Two different sensors.

And no, the gyroscope in your phone does not spin; it is solid state [wikipedia.org] .

Re:But snooped on with what? (0)

Anonymous Coward | about a month ago | (#47713813)

Tilt IS a change in velocity. In case you forgot, we are all accelerating (acceleration (noun): a change in velocity) towards the center of the earth at 9.8 meters per second per second. When you tilt your phone, the direction of that acceleration (relative to the "plane" of the phone) changes slightly.

Re:But snooped on with what? (1)

Russ1642 (1087959) | about a month ago | (#47713879)

A gyroscope can measure absolute orientation in the absence of a gravitational field.

Re:But snooped on with what? (1)

NoImNotNineVolt (832851) | about a month ago | (#47713945)

You're both partly wrong.

There are six degrees of freedom here. Three linear, and three angular. Tilting a phone is synonymous with applying angular acceleration to it. A phone existing in a tilted state relative to the ground is experiencing no angular acceleration, but its angle of tilt can be determined by combining the three linear acceleration readings into one vector (assuming the phone is stationary relative to the Earth). The word "tilt" here is ambiguous in that it could relate to either angular or linear acceleration, so you probably shouldn't be using it.

That being said, I don't know if the sensors in phones actually measure angular acceleration (which is what a gyroscope would do). When you "play a racing game" on your phone, can you fool it into thinking you're tilting the phone merely by translating (in the geometric sense) the phone from side to side without rotating it? If so, your sensor is inferring orientation from linear acceleration measurements. If not, it actually has a gyroscope (and is capable of direct angular acceleration measurements).

I repeat myself when under stress (0)

Anonymous Coward | about a month ago | (#47711877)

I repeat myself when under stress
I repeat myself when under stress
I repeat myself when under stress
I repeat

Let me guess (1)

jones_supa (887896) | about a month ago | (#47712087)

No one will ever bother exploiting this. Neither will anyone bother to exploit the red button attack [slashdot.org] or inferring audio from video recording [slashdot.org] . It's just too tricky to get these working in practice. Even with the gyroscope you get a crummy 100Hz frequency cap with terrible amounts of factors decreasing sound capture quality.

Re:Let me guess (1)

wonkey_monkey (2592601) | about a month ago | (#47712159)

Which is, of course, exactly the kind of attitude "they" would hope you'd have.

Re:Let me guess (1)

jones_supa (887896) | about a month ago | (#47712271)

It does not matter what they hope if they can't get the tricks to work.

Similar cool/scary news... (1)

RyuuzakiTetsuya (195424) | about a month ago | (#47712179)

http://petapixel.com/2014/08/0... [petapixel.com]

Good lordy.

This would be really cool if the privacy implications weren't scary. However, I can't imagine this being useful or practical wide scale. As a targeted attack, that's really scary as fuck.

Re:Similar cool/scary news... (1)

iggymanz (596061) | about a month ago | (#47712537)

been done for decades using laser on window or hard surface in room by law enforcement and others

Re:Similar cool/scary news... (1)

RyuuzakiTetsuya (195424) | about a month ago | (#47712623)

If you read the article linked, they're doing it with a cellphone camera too.

Not *great*, mind you, but possible. Thus kind of scary.

Re:Similar cool/scary news... (1)

iggymanz (596061) | about a month ago | (#47713467)

yes read the article, just saying principles not new and plenty of other ways to "bug" a room without entering. good ol' parabolic microphones can listen through walls from outside at over fifty meters range

Re:Similar cool/scary news... (0)

Anonymous Coward | about a month ago | (#47721107)

Laser microphones are old hat. What I'm doing today in my lab is "laser time of flight indirect reflected imaging". We can shoot a laser in the gap in a curtain or through a doorway and image what we can't "see" in a spatial sense. All that is required is a lambertian surface that can be scanned with a laser, including the rear side of a closed curtain. Basically we capture the finite impulse response of the scattered light off the surface, and by controlling the spatial coordinates of the step illuminant we collect a 3D matrix I of {X,Y,T}, from which we perform 3D FFT, apply a rotation matrix, reduce to 2D matrix, invert the 2D FFT to recover an I{X,Y} equivalent to the camera being at the lambertian surface.

This is actually pretty simple math, though computationally complex, and the filtering and rotation is decidedly non-trivial to get good results.

Another technique others are working on is to project two monochromatic spots onto a lambertian surface, scanning the wavelength of one or both illuminators and capturing the back-scatter. The two spots form a holographic arrangement, and the occluded interior can be reconstructed via deconvolution.

What is possible optically with phase coherent techniques far exceeds our primitive visual system which completely ignores polarisation and phase.

The fear mongering industry... (1)

Torp (199297) | about a month ago | (#47712235)

... has moved to smartphones.

Srsly (1)

Alioth (221270) | about a month ago | (#47712435)

Smartphones have actual microphones. Why use the gyro as a crude microphone when you have a perfectly functioning microphone built into the device already?

Re:Srsly (1)

Chrisq (894406) | about a month ago | (#47712497)

Smartphones have actual microphones. Why use the gyro as a crude microphone when you have a perfectly functioning microphone built into the device already?

Because its there [wikiquote.org]

Permissions. It's a kind of privalage escalation. (1)

bussdriver (620565) | about a month ago | (#47713559)

The app doesn't use your microphone; or you deny it, or whatever. So the app uses the gyro to figure out what you are saying anyway - you have no idea it can even do this because it doesn't use the microphone. 3rd parties could AUDIT and secure the software for government or corporate use--- and it would still record gyro information.

A background app could listen constantly even while other apps have the mic if it can background and use the gyro.

A hacked app with only gyro access...

Think about the story weeks ago about using video cameras detecting vibrations to hear things and what next gen phones could do with that-- similar situation (but crazy battery usage even on futuristic more powerful phones.)

Future work:
ID which person in the family is carrying the phone using the gyros?

Gyroscopic stabilization (0)

Anonymous Coward | about a month ago | (#47712649)

So, would it be possible to detect the frantic exclamations as one drops the phone and then use the gyroscopes to either stabilize or twist the phone is such a was as to prevent damage on impact. When I drop my phone I'd like to see it flip over as it fell and land safely, much like a cat does. That would be a cool and useful feature.

Dilbert! Stop working on ant milk, and ant jerky. We need cat-like phone stabilization technology. Now!

Phone permissions suck (3, Insightful)

Gothmolly (148874) | about a month ago | (#47712697)

Every app seems to want access to your full memory, location info, camera, microphone and contact list. Why does a flashlight app need all this?

I carry a phone because I have to for work, and I need something to read while on the crapper, and that's it. People who use all these fancy apps are the product, not the customer.

Re:Phone permissions suck (0)

Anonymous Coward | about a month ago | (#47713985)

My thoughts exactly, I ended up coding my own flashlight app that doesn't takes a photo or records audio of myself taking a dump. It's quite a relief.

Re:Phone permissions suck (1)

Solandri (704621) | about a month ago | (#47716149)

You can thwart this if your phone is rooted. At first I used an app which blocked apps from accessing certain features and data I didn't want them to see, like my location. But then they started to make apps crash when they were blocked this way.

I'm currently using xprivacy [xposed.info] . It generates fake data for things like location, networks, and sensors. If the app insists on getting my location and I don't give it that permission, it still gets a location. But that location is a random place in the world. Same for networks and sensor data - it'll get fake SSIDs if I don't give it network access, fake gyroscope readings if I don't give it permission to access the gyros.

Be forewarned, xprivacy is a PITA to install (uses the xposed framework) and set up (newer versions query you when an app makes a request, so you don't have to set everything blind). But after a few weeks of pain getting everything in place, it's been smooth sailing.

Re:Phone permissions suck (0)

Anonymous Coward | about a month ago | (#47717141)

While you are correct, this particular hack requires no permissions at all. They do it from a web page loaded in the browser on the phone.

Sound from Video (0)

Anonymous Coward | about a month ago | (#47714995)

After that whole 'sound from any video of sufficient quality' demonstration, this is like childs play in comparison.

Why bother (0)

Anonymous Coward | about a month ago | (#47715635)

when the mic can be turned on remotely at any time, and anything it picks up can be sent anywhere? Same with the camera, and GPS data. So called smartphones are portable spying devices.

apparently.. (0)

Anonymous Coward | about a month ago | (#47715875)

This is taking place on August 22, 2014.

Or you could just use the windows and screens (1)

WillAffleckUW (858324) | about a month ago | (#47715935)

Since a long long time ago (about 50 years now) we've been able to use nearby windows and computer monitors - even picture frames - to pick up sounds inside rooms.

Why bother with a cell phone if you're trying to get a good audio pickup?

If you need to isolate a person, it's not a bad choice, but you can also use the other signals your cell gives out or responds to for locating the person precisely, without technically "using" the phone, and thereby alerting the target.

But, hey, do it the hard way, if you must.

timing (1)

idanity (591710) | about a month ago | (#47718329)

my phone stays in my cargo pockets, When lean over in a chair to fart, they should be able to hear that fart loud and clear. now lets see them try to figure out what i ate? I also take my phone in the laveratory with me, so they should also have great opertunity to hear some of splash-logs greatest hits.
Check for New Comments
Slashdot Login

Need an Account?

Forgot your password?

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>