Slashdot Mirror
Embedding Data Signals In White Noise
Anemophilous Coward writes "ZDNet has the following article which describes a company that 'has devised a method for sending wireless signals over ordinary audio speakers so that humans can't hear them. With this same technology, radio stations can unobtrusively transmit ads, Web site URLs, or information about music and artists to in-car cell phones.'" Here is some further reading about the company, Intrasonics.
We all know this is just a cover story for the REAL secret messages in the static!
....more invisible voices urging me to do bad things.
.
.
"A microprocessor... is a terrible thing to waste." --
GeneralEmergency
Blow your mind
[A company] has devised a method for sending wireless signals over ordinary audio speakers so that humans can't hear them.
Yeah, but how much is dog insurance going for these days?
If you celebrate Xmas, befriend me (538
And life for all dog's everywhere will never be the same.
I want an in depth analysis of the Beatles' White Album immediately! Charles Manson was right!
Is this really possible? - I guess so, as long as they're only "tansmitted", and never converted into a form that can be picked up by my eyes, ears, skin, tongue, nose, ...
who will be the first artist to embed lyrics, trivia, etc in their CDs?
The truth doesn't care what I think.
Advertising everywhere... no escape. I remember reading a short sci-fi story about this many years ago. Unfortunately, it looks like somebody else read it as well...
The pr0n industry should be all over this. You can watch a movie at home with a special "doll", responding to commands...
Where can I buy stock?
"Give a woman two glasses of wine and some pad thai, and they'll agree to just about anything." the Sports Guy
I wonder if this technology could be modified to watermark the source of the signal?
But if they are saying that it is random pops and cracks how will converting it into MP3s affect it?
I guess also, how would extra noise because one has a lousy stereo do to the signal?
This is just like that time that the phone police were sending me those messages through the rings, man. Exactly the same, except different. Man.
This could be used to watermark audio, in order to try to track pirates. Granted, if they are actually using psychoacustics, then psychoacustic compression systems like MP3 or Vorbis would strip the extra data, but this would be good for basic audio.
However, I doubt this is "inaudiable" - rather I suspect it is "unobtrusive" - you would hear it, and if you know what to listen for would identify it, but you wouldn't find it objectionable in most cases.
But keep it the hell off my CDs!
www.eFax.com are spammers
Let me guess, by using the correlation of psuedo-random noise sequences summed with the signal.
-Chris
--an unbreakable toy is useful for breaking other toys--
This is an interesting idea, using psychoacoustic modeling to open a data channel in audio. The article describes some applications, and I'll certainly admit that some of them sound irritating or possibly dangerous (from a security standpoint.) Others sound better.
But not everything interesting to do with this will be done by the company involved, because it may not make good business sense or they may not have thought of it. I'd be interested in what slashdotters can think of to do with such a channel. The obvious use of embedding artist and recording information is mentioned, and I like that one a lot. It would be great to have a radio displaying those things, and to be able to scroll back and look at the last N songs. This would let you find out what that song you heard the end of was, or do a statistical analysis of a station's playlist, whatever you want.
A use that occurs to me is adding the information to advertisements so that adverisers can automate the task of making sure that they get what they pay for. Even performers could use an "ad id" check to make sure they get their voice-over royalties and the like.
Of course, voice of america and similar programs could use this right away. First they start adding this hidden content to all programming, using encrypted books, articles, or any other easily accesible source. They can then easily put a specific message with a specific key into a program that certain people can unlock. There's no entropy difference between the "real" message and the usual dummy ones to detect.
Hmmm lots of fun to be had here...
Using a seemingly innocuous message as a carrier wave for a truly useful message you don't want other people to know about is an old-news crypto technique, of course. But here's a fun, new place to apply it.
And you don't even need to seem to be doing anything funny during decoding (the message would obviously have to be enciphered; pass it in the clear and anyone who owns a cell tower between the two points can read it); build that into the phone/PDA. With the ridiculous proliferation of the damn things, no one will blink if you receive a call, chat for a few minutes, and then tap a few buttons. For all they know, you're sending an SMS, even if you are entering your passphrase.
All it really takes to do 3DES or Blowfish in software in a reasonable period of time is a StrongARM or similar (my Newton's got one, you cell phone must), though you'd get far better performance doing it in hardware. (Watch out for escrow, though!)
Do you have a
Ignore the messages embedded in this whitenoise.
You will Loooooooove Microsoft
You will Haaaaaaaate Open Source
Linux is eeeeeeeevil
War on Iraq is goooooooood
"As far as I'm concerned, I prefer silent vice to ostentatious virtue." ~A. Einstein
Hm, could someone send a mass-broadcast virus this way?
Trolls lurk everywhere. Mod them down.
I find it sad that everytime a new technology such as this is developed, the first instinct of the marketing people tasked with selling it is to figure out a way to make it push ads into my perceptual environment, almost guaranteeing an initial cynical reaction..
Except for the fact that the article SPECIFICALLY states that it doesn't work on the same principle as dog whistles as the sound couldn't be transmitted through ordinary speakers.
"I can hear it, can't you?"
-dameron
Will my pet Hamster Fred, freak out ? Inaudable usually means higher sound frequencies. If it is audible but comes off as white noise
Nah. If you've got an infinite amount of data that's random (not saying pi is), then probability that any given string exists within it tends towards one. But the probability of any given string you select *being* that string tends toward zero. So you'd probably spend a very long time (not a finite time, even) looking for the offset. And then the person at the other end needs a function that generates the same "random" data, which'll take them an infinitely long time, too, if your offset is an infinite.
Now I don't know shit about maths (tends toward might actually be "1", for instance) but that's kind of how it seems intuitively. If someone's going to rip me a new one, at least answer the guys question with your superior knowledge.
Can someone please tell me how to remove this pop-up ad that's sticking out of my ear? Everytime I think I've gotten rid of it it comes back.
The Radio Data System has been around for ages and it allows precisely that: transmitting extra information with normal radio signals... Because it works by putting digital signal into inaudibile frequency, it should do exactly the same, as long as speakers have any response at 20-40Khz.
"With this same technology, radio stations can unobtrusively transmit ads, Web site URLs, or information about music and artists..
Humans tend to filter out what they don't want to hear, especially the pop, fizz and hum of white noise."
So if I understand this correctly, the technology can transmit advertisements, spam, and pop music completely unheard by the human ear by disgusing them as advertisements, spam, and pop music?
Shameless plug for my photos on Flickr
A long time back, I was reading some hardware reviews on sound cards. 1 of the reviews mentioned a card that was rather nominal, but came with a special set of headphones. In about a page of article, it glanced over the soundcard and then went into rave reviews of the headphones, which apparently used "white noise" to block audible frequencies except for the music/etc coming through the headphones.
My guess would be that this could be used to create a signal, and block it from human perception but perhaps still allow it to be picked up be electronics.
One question I have though, if they're making such advanced uses of white-noise technology now, and these headphones/soundcards came out over a year ago, why haven't I seen rave reviews on the technology and white-noise headphones available at every radio shack? From what I remember, including the price of the card the 'phones weren't that expensive.
"History doesn't repeat itself, but it does rhyme." Mark Twain
even though the average human ear cannot hear these messages, it is very likely that we will see a large variety of animals go cracy
Well the article says that they spread the encoded data throughout the spectrum, not that they place it frequencies that we normally can't hear. My understanding of animal hearing is that they hear frequencies that we can't. Assuming that this is the case, then animals shouldn't be too affected, if at all. I'm definitely not a scientist so I may be off base here, but my first thought too was that they were using inaudible frequencies, but the article seems to suggest that they don't. Which would make sense since your average oem car speaker would have a pretty tough time reproducing sound at frequencies high enough.
It'll get published by 2600 and then every kiddie will be encoding messages and sending them out through their little radio shack fm transmitters as mommy and daddy roll down the highway, making the technology useless (or even more useless)
isn't white noise like britany spears, nsync, backstreet boys,etc?
oh wait...you mean the OTHER white noise
nbfn
Does the company's technology work on the dog whistle principle, using sound waves that are below the threshold of human hearing? No. If it did, you couldn't send the signals over standard audio speakers. Instead, the technology revolves around what's called psycho-acoustic masking.
However, I would still like to have someone explain how this will work in reality. What kind of performance will be required to catch and decode these messages. It does not seem like a very light task.
Others have already observed that it's not a frequency thing, but let me expand on that.
Frequencies are already optimized for human hearing, and it's not usually possible to send, say, a 40,000Hz signal on most anything you can think of, analog or digital. Standard phones have a bandwidth of something like 3 K Hz. CDs of course top out around 20,000Hz, give or take a bit. (It's not a perfect cutoff at 22050, it's a curve, so there isn't quite a point you can say is "the limit".) I don't know for certain but I'd bet FM can't transmit those frequencies and be compliant with FCC regulations. (Of course the tech could do it in theory, but the radio station may have to leave their allocated frequencies to do it; I don't know for certain.) AM could do it in theory but based on the low quality of the signal I hypothesize that something is preventing high frequencies from getting through.
Finally, the coup de grace is that our speakers are optimized for human hearing, pretty much no matter what. Covering the bases from 20Hz - 20,000Hz is a hard enough problem without pushing the required range up another couple of octaves.
In fact, what the company is proposing seems to be in some sense the inverse of MP3 coding. MP3 coding strips the signal of things that you can't hear through by analysing what is psychoacoustically masked in the original signal. The MP3 encoding process can then focus on just the parts of the signal you do hear, which is obviously going to require less space, except in some pathological cases where the whole sound is perceivable (like a pure sine wave tone).
From what I understand of the marketing, the part of the signal that an MP3 encoder strips out is exactly where they would place their data. They can stick any data they want in there and we just plain won't hear it, but a computer+microphone doesn't have this problem.
Interesting corrolary: The time frame this will work in is limited, as digital transmission usually uses compressed audio, and the act of compressing the audio will preferentially eliminate this data. (Or does digital radio transmit an uncompressed stream?) They'd better get marketing this now, so that there's an installed base and they can try to later create receivers that will re-add their signal on the receiving side. Of course, if all anybody is using this for is advertising, I can't imagine we're going to go out of our way to buy "Advertising Enabled!" digital radio receivers.
...sending wireless signals over ordinary audio speakers so that humans can't hear them....
Is it just me, or does embedding data in white noise "sound" like it's already happend? Every time I pick up the phone when someone else is using the line for a dial-up connection, I am abruptly reminded of the transmission of data using seemingly random noise....
$ # Patent pending...
$ bzip2 -c </lib/libc.so.6 >/dev/audio
And how is this diffrent from steganography + a pair of 2,400 buad modems?
Besides, elephants have been doing this for millenia (with their feet instead of over their THX system).
moto411.com
A hot new startup in California has announced a technology for encoding color information in black-and-white television broadcasts. The extra signal is encoded such that black-and-white receivers don't notice it, using a proprietary technique referred to as a ``subcarrier''. Millions of Slashdot kiddies are smitten with awe.
Isnt this the story of the Movie The Ring?
One of the things that came with the cuecat kit was a rca cable that was meant to go from your computer to your sound card. Apparently, while watching TV they'd embed a signal into the audio that the cuecat software would pick up, and it would take you to their site.
Since one of our local channels was owned by the Belo corp (who owned a LOT of Digital Convergence stock), they pushed it HEAVILY, and embedded URLs in the news program.
So, nothing new.
"Sometimes a woman is a kind of religion, she can save your soul & set you free from all your sins" - Bad Examples
The company has discovered that by adding static and degrading your music quality they can send data over sound waves! Woohoo! You know, like a modem! Err, actually it's only half a modem, it's only one-way! I'm going to use another exclaimation point!
After a pretty thorough search I was unable to find a data rate for this new top sekret modem technology. I'll confidently wager it's going to be between 10 and 100 baud, as in a 0.01K modem or 0.1K modem.
-
- - You can't take something off the Internet! That's like trying to take pee out of a swimming pool.
Yes, except that the little girl tells you to buy Windows XP.
Finding God in a Dog
Embedding signals leads to noise with cumulants different from zero for different times, and thus the noise is not white anymore, it becomes colored.
Your display probably isn't true white either; it's made of tiny red, green, and blue dots. This article is about perceptually hiding information in audio that humans hear as "white" but that machines can pick apart. It's steganography.
Will I retire or break 10K?
Wouldn't something like this easily fall into it? It'd be interesting to do a study to see if people responded to stuff transmitted just on the fringe of their hearing range.
/; insert microsoft XP CD.
This kinda scares me a bit. I feel the sudden urge to rm -rf
The next site to slashdot will be ready soon, but subscribers can beat the rush and start slashdotting it early!
The one thing I can't stand about cell phones is the paucity of advertisements. It's really difficult, I go into withdrawal sometimes while on the phone, because there are no ads telling me what to buy. It's almost like being in a forest or cave! Thank goodness they are finding new ways to infiltrate the mental environment.
In europe, there exists a similar technology called RDS for "Radio Data System". It's on the air for about 10 years now and allows for these cool features since then:
- Show the Station name in your radio display
- Show what's playing
- Certain stations are transmitted over several frequencies. RDS knows the alternative frequencies of your stations and automatically switches to the best frequency
I remember a broadcast I watched on TV half a year ago. It was how to use infrasonic soundwaves to create atmosphere in cinemas. With all that THX and Dolby-stuff around two sound engineers who had worked together as close friends suddenly started to fight each other over something in the sound studio. Afterwards they realized that their mood had been seriously affected by unherable infrasonic soundwaves they used for the movie they were working on. The reporter said that this is used in pretty much every new movie to create an atmosphere of fear or rage among the viewers. I wonder how far this could be taken...
There also seems to be a strange deep sound around in Europe. Nobody, not even the scientists measuring some American farts deep into the bavarian woods, was able to determine the source of that sound yet. Over 1000 people in Europe seem to be affected by that. They even made some conlusions about global soundwaves created by military sonar and stuff. Pretty scary...
So I'd like to know about every stupid signal I will get, purely encoded data or else. I'm not a technophile but those stuff scares me because I won't be able to switch off the router or TV anymore...
0 001 11 1
Radio is limited to about 16 Khz on FM. Most of us can easily hear that. So they up the noise floor a little to make room for ads?
Does this mean we get fewer audio ads in trade for the lower quality?
Pipe your radio through lame, output through sound card, information gone all done!
Blogging because I can...
It seems like they're using psychoacoustic masking -- which really isn't a bad idea, as it won't change the perceptual SNR (unlike the spread-spectrum white noise espoused here). Psychoacoustic masking carefully removes bits of audio information that we would be unlikely to hear anyway. Dolby Digital gets about a maximum compression about 16:1, which would do for cheap-o speakers, but it would probably be placed more around 5:1 so those with nice stereos wouldn't hear the difference. MP3 uses similar masking compression ideas, and a 128k mono bitstream gets a compression of about 5.5:1.
Let's assume some rockin' speakers with a 22kHz rolloff, and a great FM reception with 96dB Signal-to-Noise (and a very quiet listening room). That's approximately 16 bits at 44kHz or 704Kb/s of information. That has to carry both the audio and the data signal. The data signal would have to be mono, since most toys & cell phones don't have two ears (err, microphones). Now, you have a 704Kb/s bandwidth in which you only need about 1/5 = 140Kb/sec for good audio, leaving you with a theoretical maximum of 563Kb/s left for data. Put in some forward error correction and packet and coding and other overhead and you'll probably get something more akin to 200Kb/s.
But wait! Let's assume a car with some poor tweeters with a 15kHz rolloff, and poor FM reception with 65dB Signal-to-Noise with road noise added in. That's approximately 11 bits at 7.5kHz or 82Kb/s of total information. Ooops! You've exceeded your channel capacity by almost 2x, and you'll pretty much get a big fat zero data bits.
So, the makers have make a tradeoff ->
1) Low data rates: significantly less than ~200kB/sec to accomodate cheap stereos but retain audio quality.
2) Poor audio quality: significantly less than ~140kB/sec to accomodate higher data rates or cheap stereos.
3) Lose functionality on cheap stereos: but retain both good data rates and quality audio for those who can receive it.
My guess is that they'll just go to something tiny like 500b/s, in order to reach the most market share. Even at that rate, a text ad would come through right quick.
I can just see the next Furby craze, now they get instructions (programming??) from the TV!
Anyone know the max bandwith and SNR of NTSC audio?
-- Heisenberg might have slept here.
"Buy the CD... Twenty-five dollars is cheap... Go ahead... You know you want to... Uh huh... Buy two copies - you need a legal backup... Come on honey... Hilary needs a new Cadillac... Only thieves copy music... Copy music and go to jail... Turn in your neighbors who copy music... Oh yeah, baby..."
Sigs are bad for your health.
Another method for the delivery of spam. Guess we'll have to start work on the anti-white (black) noise generator.
Check out the Portable People Meter from Arbitron. It can recognize subaudible watermarks in music including over radio, Musak, and even some streaming audio compressions. Arbitron uses it for ratings purposes.
Of course, then there is IBOC from Ibiquity which is an on-channel digital enhacement for AM and FM signals, part of which could be used for datacasting, as part of most DTV signals will.
MP3 and OGG are essentially based on filtering OUT audio that would be not be heard due to psychoacoustic masking :) ..... so.....guess what? :)
:) OSS to the rescue fighting unwanted ads again!
Just chain a realtime OGG encoder to the incoming music stream and it should strip the info! HEHE
Find a job you like and you will never work a day in your life.
If the CD digitally stores 44,100 samples a second, why isn't 22,050 an absolute cutoff? I would imagine it would HAVE to be, since it's not physically possible to make the wave go up-down-up-down any faster than 1/2 the sampling rate.)
Don't label something "offtopic" unless you know the topic well enough to tell what's on topic.
Imagine, the universe is filled with spam. Ahhhhh!
Friends don't help friends install M$ junk.
This technology could encode bits in it that tell recording devices they can't record. I know what you're thinking: use old technology. But what if you record this and then all the new cd players detect this flag? Any other thoughts on this?
This is Push technology, so you could do any kind of push you do with channels now you can do with this. Here a couple of idea's I had:
Song Titles - How about ID tags so you can actually see the exact title of the song your listening to? You can keep a 10 song list like caller ID. You can see the last few songs you listened to. For advertisers you would keep one text line scrolling on the display.
Second Audio Channel - A secondary program at AM quality on an FM station. Kinda like how HDTV can have up to 6 lower quality channels.
Radio for the Death - Close Captioning of radio lyrics. 'Nuff said.
Emergency Broadcast Technolgy - Give both a readable text warning and GPS cordinates of pending danger.
Exact Station Date and Time - Isn't that what you really want half the time you turn on the radio anyway.
Weather and News Broadcasts - Get the local or national news in an instant.
Automatic Request line # - Never have to listen for those damn # to call while your driving.
Possible interactivity - Broadcast a survey to a cellphone, you log your answers, and then you transmit your results.
You say things that offend me and I can deal with it. Can you?
Embedding data signals in white noise is the easy part. It's getting them out that they haven't figured out yet.
Considering that a modem carrier sounds like noise anyway, it seems that this was only coming soon to a mall sound system near you. But does this mean that the bells I hear in Robinson's May while my wife is shopping will be carrying other data? =)
This sig no verb.
Just FYI: The audio bandwith of stereo FM radio is about 18kHz or less. The stereo pilot signal is at 19kHz, and the Left-Right signal is centered at 38kHz. So the radio receiver is going to need some pretty good filters to produce the full audio bandwidth, and eliminate the 19kHz pilot.
AM radio stations are only 10kHz apart so you are limited to less then 5kHz of clear audio bandwidth.
This web page has some good info. It shows only 15kHz of audio bandwidth for FM stereo which is probably typical, but that is not a limitation of the specification.
This is really interesting from a watermarking perspective. Think of any Industrial song that makes use of sampled static. Just because it is noise, does not mean it gets filtered out by the codec. The noise is part of the signal.
You may expect steganographic stuff to get munched, but what if the watermarking software identifies noisy passages and replaces them with apparently random "noise" that is your watermark? Unless you knew about it and delberately munged "noise", your codec would try its best to *preserve* that signal. Filter too much of the timbre (aka noisiness) out of your signal and it starts sounding lifeless.
Xix.
"Everything is adjustable, provided you have the right tools"
I have always liked when I put a CD into my player and the name of the song and the artist comes up on the LCD screen. Wouldn't the use of the "White noise effect" allow radio stations to transmit this information to radios for the same display purpose?
*song is ending* "Damn, I love that song, I wish I could remember who sings it. Maybe the DJ will say the song title before the next song comes on." Oh wait, DJ's don't do that anymore...Just show it to me on the LCD!
-Krnl
http://krnlpanic.com
I hadn't considered the size of the offset. And it seems blindingly obvious now you mention it. Ta.