Slashdot Mirror
MS Researchers Develop Acoustic Data Transfer System For Phones
angry tapir writes "Smartphones that support NFC have been making their way onto the market, but many handsets still don't support the wireless technology. As an alternative, Microsoft researchers have prototyped a system that instead uses a phone's microphone and speaker to transmit and receive data. The P2P data transfer system uses a novel technique of 'self-jamming' to stop nefarious third parties from monitoring transfers, and the researchers believe it's more secure than standard NFC communications. No word on whether it sounds like the squeal of a 56k modem."
It's amazing what comes back as "new developments"
I don't have time to make a sig
So they reinvented chirp.io ?
"modem"
Sheesh, evil *and* a jerk. -- Jade
Here we go again !! Whitesnake !!
acoustic coupler?
next M$ discovers moving pictures
...did they patent it yet?
They want their acoustic coupling back...
GREETINGS PROFESSOR FALKEN.
The acoustic coupler...
http://en.wikipedia.org/wiki/Acoustic_coupler
ahahhahahahahahahahahahahahahahahahahahah laMer$
Because when I think of secure, reliable communications I can trust, it's Microsoft I think of first.
http://www.computerworld.com/s/article/9217790/Sound_based_system_promises_chipless_NFC_now
You can already transfer music between phones like this, but it's quite lossy depending on the quality of your speaker.
"Mr. Watson, come here. I need you."
Have gnu, will travel.
I guess ms might be out of ideas. If you can't keep up with 2013, you might as well attempt to keep up with 1970.
Researchers will find an alternative for phones that don't have either Bluetooth, NFC nor acoustic data sharing.
Linux is for people who don't mind RTFM.
As soon as it is released, I bet a load of people could fool this crap just by whistling a few notes.
Near Field Communication
Unlike a modem that requires a carrier tone, two acoustic devices that need to send a couple frames of data (such as a Diffie-Hellman exchange) could easily send and receive the data with a few bursts. DACs and ADCs are good enough to be able to discern the encoded static, find errors and correct them, and pass the decoded packets along. This wouldn't be fast, but it would be good enough for creating a shared secret or just validating each other's public keys so future communications can be reliability secured without need of a CA.
Wow, return of the acoustic modem. That really is a trip back in time. Was cutting-edge technology, back in the era of blinking-light consoles, when telephones were hardwired into the wall.
Ah, nostalgia for the tech of yore.
http://www.geoffreylandis.com
It's almost like they're modulating a signal and then demodulating it. I wonder if there's a name for this sort of thing.
I recall some technology that transmitted data over POTS lines with sound. What the fax is that called again?
First this is a wonderful idea so I don't want to put it down as a useful contribution to the low bandwidth limited distance problem for comunications. Where the authors seem to go south here is the huge time they devote in the article to touting that NFC has no physical security and their system does via "jamSecure". Unless I'm missing something there's no reason, other than changing the standard, that radio based NFC could not also implement JamSecure and even do it better. The idea of JamSecure is that both ends of the communitcation channel transmit at the same time, anyone listening in hears the sum. If one of the emitters is sending simply random noise then the sum is randomized. Yet because the receiver knows what they are emitting they can subtract it out. Don't see why NFC cant do that. Also I don't see why having two (or more) microphones in different locations on an eaves dropper doesn't ruin the addition the encryption is relying on. At least with NFC you can have the transmitters be spatially diverse too, with sound that's harder.
But for very close by communications using existing tech, why not use the screen and the camera? Each phone looks at the others screen and reads it. bandwith becomes the screen refreshrate time the number of resolvable pixels. Presumably at a meter or so that should be close to or better than sound in band width.
Some drink at the fountain of knowledge. Others just gargle.
"Dear aunt, let's set so double the killer delete select all"... that is the code for activating the system...
If you keep throwing chairs, one day you'll break windows....
Everyone's already noticed that this is just an acoustically coupled modem setup. But this is better. Put the receiver of one by the speaker of the other and vice-versa. Now you've got two phones literally coupling, like 69, soixante-neuf, right there on the table at Starbucks.
I am not a crackpot.
http://blog.ncf.ca/wp-content/uploads/2011/09/apple_apple2_acoustic-coupler_1.jpg
I hope this thing works faster than 300 baud
The only drawback is that you need this aftermarket hardware to plug it into the cradle.
It was invented at MS India.
It IS a new original idea for the persons involved.
The persons involved are probably younger than the last produced acoustic couplers, possibly modems, and even then there were not many of those in India when they existed, and there was no internet to learn about them either. Most information about them only exists printed on dead trees.
I have a Polar bicycle computer that is over 5 years old. To program some of the settings, the unit must receive the data via sounds from the desktop's speakers.
...Play a game?
The problem with quotes on the internet, is that nobody bothers to check their veracity. -- Abraham Lincoln
I had this idea a while ago as a way to order from a drive-through window. Create your order on the McApp, hold the phone up to the drive-through speaker box and it squawks the order through in a second. Pops up on the operator's screen, they read out the price and you go through without having to yell "I said NO ONIONS" over and over again. When I researched it, turned out there was an Apple patent covering exactly that use case so I gave up on it. I wonder if the MS researchers will run into issues with that bit of IP.
Don't all these devices have bluetooth transceivers already?
My LG front loader has a way to send diagnostic data to the factory; you hold the phone's mic to the washer and it sends data to the factory (which you presumably have to call first).
By jove! They've invented acoustic coupled modems!
http://en.wikipedia.org/wiki/Acoustic_coupler
Someone has been watching War Games again!
Near field communication always sounded like a bad idea. This tells me that it is worse than I imagined.
If an acoustic coupling method/process is an improvement then the security of near field communication is in a sad state indeed.
OR this is a sad attempt to promote a patented standard to extract more $$ from hither and yon...
Truth is stranger than fiction, but it is because Fiction is obliged to stick to possibilities; Truth isn't. Mark Twain.
This reminds me of the old TV remote controls that had a set of tuning forks. Press a button for channel up and a hammer hits a fork. Channel down is a different fork.
Sounds like this shit called Illiri http://illiri.com/api.html
the speaker systems the drive-through speaker systems need to be better for that to work. Some of them are real bad.
Cell69, because that's what your phones do to make it work.
let's see, 100 comments, less than 10 that actually discuss the article.
Welcome to Slashdot, News for Nerds, 90% garbage comments,
This tech is old, they are called "ringtones" by others. They transfer information quite reliably telling others, "The phone is owned by a total *sshole" when certain things play with near 100% accuracy. MS is probably going to use this to patent ringtones, and start suing android vendors.....
-Charlie
http://www.penny-arcade.com/comic/2009/7/31
[quote] Microsoft researchers have prototyped a system that instead uses a phone's microphone and speaker to transmit and receive data. [/quote]
You mean Microsoft bought a company whose researchers have prototyped a system that instead uses a phone's microphone and speaker to transmit and receive data.
Microsoft has bought every innovation they ever made all the way back to DOS.
I could imagine the ambient light sensor and a flickering backlight being used to send messages in a similar (and perhaps less annoying way) too.
Another flaw is that Jam Secure isn't either in the audio or the hypothesized radio implementation.
A signal being jammed by another of comparable strength and nontrivial spacial separation can be received by TWO or more microphones, or antennas, also with nontrivial separation, and the signals sorted out in postprocessing (at "line speed").
This is how MIMO works: Two (or more) transmitting antennas send different signals, two or more receiving antennas receive sums of them - which differ because each antenna "hears" different time (phase) offsets due to the slightly different delays in the paths to each antenna (or microphone). The receiving end sorts out the signals. In MIMO this is used to send, on the same bandwidth, up to N times the bandwidth of a single antenna pair (where N is the number of antennas on the fewer-antenna device). But it can also be used to sort out a particular transmitter from spacially separated devices.
When a cell tower does it, using N antennas to sort out a desired cell phone from N-1 interfering transmitters (and maybe doing it N times to hear N cellphones at once) it's called things like "steerable null". When a human does it, to sort out one speaker in a crowd, it's called "The Coctail Party Factor".
It should not be difficult at all to do the same with a pair or more of microphones and a good DSP or a fast processor. For audio, where wavelengths are measured in one to two digits of inches, the separation of a pair of handheld devices (even if nearly on top of each other) should be more than adequate to do MIMO tricks successfully and without obvious eavesdropping equipment rigs.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
You're not transmitting a huge amount of data. 75 baud would be plenty.
What's this "hard drive" you keep going on about? And what's this about 56k? No modem can go faster than 110 baud.
This is neat, although (as others have pointed out) not exactly a new idea. In a world where all cell phones have a speaker and microphone under software control (and in most cases, an accelerometer, supporting a "clink to sync" mechanism for short-term pairing), how did the concept of NFC, i.e. a separate antenna*, receiver chip and extremely application-specific software stack, ever get off the ground?
The typical adult human cannot hear frequencies above about 15KHz (a child can - anyone remember that flyback whine from CRT televisions? - but only if the amplitude is high enough), whereas the phone mic samples at typically 22KHz or 44KHz. The antialiasing filters on them are invariably shit, so ultrasonics on the 22KHz device will just alias into the passband and make your decoding job almost easier.
* (And if we are going to put a big, low-frequency RFID antenna in the phone anyway, where, oh where, is my builtin induction charging support?)
Caveat Emptor is not a business model.
goes around?
Camera/Screen combo is already used by banks.
I've seen 2 variants:
- A german bank I did use shortly did display a blink-ing barcode type of transmission on their e-banking webpage, and the pocket chip TAN (a calculator-like challenge/response/pin device) has matching photodetector for each blinking line on its back. You cover the blinking barcode with the device and wait for it to read/decrypt the signal.
- A swiss bank I'm using uses "Photo TAN": the e-banking website provides as 2D barcode (something like a more complex color QR-code), which is decoded by an application (iOS or Android) running on a cam equipped device (smartphone, tablet, etc.).
In both case, eaves droping is made modre difficult: unlike sound or RF which are broadcasted in all directions, images are only visible to the person in front of the screen and are pretty trivial to hide : just cover them (and in the first case, that's actually part of the way the device works: you HAVE to cover the code with the body of the device so the photodetector and the code align).
If you want to eavesdrop, you'll either need tempest-like sigint, or you need to compromise the desktop computer to the point where you can access the screen.
And even then, the system still resists man-in-the-middle attacks, etc. because the device (chipTAN or smartphone/tablet) displays which transaction it is actually validating before producing the TAN code.
(Only the photoTAN scheme has a possible security failure, if the smartphone gets remotely hacked, and the secret key stolen, and the key password key-logged).
"Sufficiently advanced satire is indistinguishable from reality." - [Tips: 1DrYakQDKCQ6y52z6QbnkxHXAocMZJE61o ]
Looks like someone else won the race to the patent office (http://www.margento.com/technology/data-over-voice) by a few years. I wonder if they'll just buy them now.