Spooks' secret TEMPEST-busting tech reinvented by US student A mysterious secret technology, apparently in use by the British intelligence services in an undisclosed role, has been reinvented by a graduate student in America. Full details of the working principles are now available. BAE Systems' wireless through-hull comms demo at Farnborough 2010. Works through glass, too. Tristan …
Acoustic comms have been about for years. This is just the transmission of sound through a dense medium. Same as Sonar and acoustic modems, just through metal and at higher frequencies (lowering range and increasing bandwidth).
The principle behind this 'invention' is actually a well-known and well-documented pain in the ass for people doing underwater positioning; we have to make sure that acoustic positioning beacons are kept a decent distance away from, say, big steel structures we're putting onto the seabed. If not, the sound travels through the steel as well as through the water and gives screwed up results (the speed of sound through steel is much faster than it's speed in water, so you get 2 identical returns at different times).
So if he tried to patent it there are at least a dozen companies out there ready to strike down that patent. Not to mention various governments getting pissed off with him and stopping the patent being granted.
@"but what is the detection range of sonar"
Ask a Whale. They can hear sounds from over a 1000 miles away, but with all the noise in the oceans these days, that range is usually down to a few hundred miles.
@"More worrying is what 50W of ultrasound does to a solid structure over days/years"
More worrying is what it does to Whales and other marine life!
What carries for distances underwater is, I believe, the LOW end of the sonic 'spectrum" -- the bass notes of the whales' song. High-frequency sound, as with high-frequency light, is more easily scattered and dissipated in both air and water. The U.S. submarine service, for instance, uses Extremely Low Frequency (ELF) radio transmitters to send signals to vessels that remain on-station and submerged for weeks at a time.
The article here refers to "ultrasonic vibrations", so we're talking high-frequency, short-penetration waves (which probably explains the statement that "(i)t seems certain that performance could be traded for range," that is; that the frequency could be lowered, allowing greater penetration while lowering the amount of data that could be carried by those fewer cycles per second.
So, I suspect that, unless the whales are snuggling up to the sides of a nuclear "boomer", they're reasonably safe.
I was thinking the same thing. Long-distance sub-sea communication is probably INFRAsonic in nature rather than ULTRAsonic. Besides, large animals such as whales are more naturally capable of producing infrasound. On land, elephants use the technique as well, IIRC, sending infrasound along the ground.
Sound doesn't travel very far at high frequencies, too much is lost. Underwater at the upper range of human hearing nothing over a couple of kilometers would be detectable no matter how "loud" it was. Half the energy would be lost in the first km at 20,000Hz and then half again in the next km.
At the lower end of human hearing 20Hz it would travel hundreds of km for a whale song level of volume.
Whales use low frequencies which travel far. Ultrasound does not travel far and will be reflected or absorbed within a short distance.
The 50W is not getting radiated out into the ocean if it is being harvested for use by a device. There will be leakage none the less. Likely that amount of ultrasound within a room would be quite obvious to some sensors.
You all suggest that this kind of tech is exclusively gonna be used in marine-vessels.
You still need that outside power-source to drive those 50W. If it's going to be used it'll be on the inside of submarines etc. to stop cutting holes i/t various compartments for data- and power-cables.
In space however... there are no whales :-)
You could interlink various space-crafts'electrical and data signals just by physical contact with these transducers. It would make coupling the various space-modules on the e.g. ISS much easier (and quicker).
Anyway all these type of "inventions" still need some electrical source. The main global problem today is how we're gonna make that source (without dependency on fossilized resources)? Ppl should concentrate more on searching for a viable long-term energy-solution instead of these electric-or electronic gimmicks.
And then we should prioritize space travel and off-world colonization as this planet is becoming filled up with human trash.
Just MHO.
Sonar is a strange thing.
If you're in a submarine and you send out a sonar pulse everything else in the water knows where you are.
a large proportion of Sonar arrays are passive (no sound produced) rather than active, they listen for other noises, while they have their limitations (you get echo's and left and right get confused, distance is also not as good) they would probably pick up pretty much anything that makes _any_ noise.
active sonar is no good for war/stealth situations, and is not the cause of whales going ditzy, that is down to the trials of infrasound for communication (travels faster in water than in radio in air, much faster messaging) and I believe that the trials by both US/UK have had to change to compensate for the known whale related issues.
The latest passive acoustic (and sometimes-passive acoustic) sensors (in use with both the military and commercial sectors) are great- they can be used to, say, detect intruders- and these can even determine the TYPE of the intruder (ROV, surface-fed Diver, scuba diver, etc) and give a very accurate range and position. I've seen them in action and they are seriously impressive bits of kit.
Acoustic comms don't travel faster in water than radio does in air- in fact they're orders of magnitude slower (somewhere in the region of 1500m/s, compared to 300,000,000m/s for radio in air.
What an acoustic signal DOES do, however, is actually travel through the water. Radio has a horrendously short range in water- basically zero unless you want to get into some serious maths, then you can squeeze out some range in certain circumstances- whereas ELF acoustic comms can propagate relatively slowly (though still faster than most jets- the speed of sound in water is ~4x the speed of sound in air, so a plane would have to go Mach 5 to overtake this signal!) through hundreds or thousands of miles of water.
And the higher the frequency, the shorter the range but the more information you can encode onto the wave in a given time (for example, at ~30kHz using RPSK you can encode >1kbit/s and transfer it 3km through a water column). ELF would be useful for sending more code at a slow manual speed, but over hundreds or thousands of miles. It's also harder for equipment to get a good quality lock on the source of an ELF signal unless you've got very complicated, specialised equipment or multiple widely-spaced listening points. Whereas a 35kHz signal can be pinpointed to a few mm from kilometers away (my personal record with this kit is a 4mm window for error at 3000m- using just sound!) using just the one, man-portable transceiver.
What happens if you put a non-ferrous material that interrupts the vibrations of the 'metal conductor'?
I mean starting with wrapping the metal at a point somewhere in the middle with a soft rubber wrap, or
using a rubber gasket along with either a non metallic connector or a metal with different harmonics.
I would also suspect that if you used plumbers tape on the threading that too would cause interference.
I'm sure there are other ideas on how you could also defeat this...
transformer coupling → #
Developed as a joint project by NASA and IIRC the University of North Wales.
The target was looking at some way to handle problems like the rotating joint on the solar arrays of the ISS needing both high power and telemetry channels.
Off hand they were talking of of power transfer in the 100Kw range and data rates in the Mbs range (both with significant capacity for improvement with the high efficiency of transformer coupling.
Transformer coupling only works for a distance of a few wavelengths. Once you are in the far field, you get a wave which needs an electrical component which cannot form in a conductive material.
Further more Eddy currents will greatly attenuate the magnetic fields at high frequencies long before that.
If you think about the problem, the answer is obvious:
unable to use electrical connection
unable to use magnetic connection
unable to use radio waves
Just about all that leaves is audio and mechanical movement. Ramp up the data-rate and reduce detectability/annoyance by using ultra-sonics, using the structure as a mechanical transfer.
Simple really.
Just wait for the foam rubber based (vibration damping) faraday cages to appear.
the people behind this (http://www.sciencedaily.com/videos/2007/0409-metal_rubber.htm) rubbing their hands with glee- it's conductive and it's rubber, so it'd work as (or at least augment) a Faraday cage AND damp out high frequency noise!
With the right business guys, this could make them very wealthy.
Using a combination of Ruby LASER rod and probably piezoelectric crystal* it created a wave they were calling a hypersonic sound (because it traveled faster** than sound in any media that transmitted it.) They wanted to use the effect for inside the egg egg scramblers and death rays and so forth.
Last I heard of it was being used as the plot device for some spy kills spy TV episodes (Man from UNCLE used it once.)
As far as it transmitting power; the little goody that powers your florescent light in the back of your laptop screen works much the same way .
*firing into a Sapphire? I think I remember that.
**Obviously not a common compression wave, possibly the same sort of tech they worked up down in Sarasota Florida in the 70s.
I don't have the slightest were I found out about all this, I remember there used to be something called a library and vaguely remember reading something there.
if I wanted to stop anything being transmitted via power cables I would make an isolated supply
3 options spring to mind
option 1
take the mains, rectify & smooth and filter it
then feed the output of that into an internal inverter
if I wanted to be really nasty to anyone trying to pick anything up off the mains I would make it a square wave driven inverter (which should throw out lots of hash)
option 2
M-G Set (http://en.wikipedia.org/wiki/Motor-generator)
basicly an electric motor coupled to a generator.
these were used heavily in military aviation for creating the higher voltages used in the radio systems but are more than capable of being scaled up to run larger loads.
option 3
Diesel generator in a secure compound
ferrite will only filter out high frequency stuff (otherwise you couldnt use it on 50hz mains) the hardest thing to filter would be a low speed data transfer being done increasing and decreasing current draw on the mains at (maximum transfer speed would be 100bps and it would be very susceptible to interference) but it would get through ferrite untouched and likely get through an inverter since that would pass its increased load on to the outside world
The mog-gens in aircraft are used to produce 3-phase 400 Hz power for avionics. That's the relentless whine one heard in cold war surveillance aircraft. We also used them in the ground support facilities for the same purpose -- 60 Hz single phase in, 400 Hz 3 phase out. Big, heavy, and not particularly. efficient
Mo-gens are also used as a type of UPS -- a big heavy flywheel keeps the generator part spinning while a diesel generator starts up.
Way way back, the spooks discovered how to read decrypted telex traffic in foreign embassies by monitoring electrical noise on the power lines going in to the code room. This varied minutely depending on what character was being set up to print next. I imagine that power lines have been carefully shielded and decoupled ever since this came to light, more than 50 years ago now.
> Get power in and out?
High quality filters. Here's the first relevant hit from googling for screened room mains filter:
www.tesch.de/english/download-fu/mainsfilter-shieldedrooms/a11x73.pdf
100dB+ of attenuation will stop your Ethernet over mains plug from working.
Note the leakage currents though; don't forget your hard-wired secondary earthing.
"The question is, did we patent it? LOL" .... Anomalous Cowturd Posted Thursday 10th March 2011 13:45 GMT
I take it that is sarcasm, or is it irony, AC? :-) Patent spooky technology? I don't think so. One just moves the goalposts every now and then and as one feels like. A little trick learnt/codified in Bletchley huts before most anyone presently working in such fields were even born.
And do you imagine that such works as were leading the field then, stopped after the major hostilities or went deep underground with a cover that they disbanded and work discontinued.?
I was hoping someone had finally managed to prove the existence of axions and at the same time put it to good engineering use [which would be - pass light through magnetic field on side A, some photons turn into axions, pass through wall, where another magnetic field turns some back into photons for detection]. That would have been something, even at 330 bps.
Still pretty good though.
Nah! they keep an eye on patent applications and then issue a confiscation order on anything they like. The owner is given nothing and told he will go to jail if he even talks about it.
Then his idea/tech/solution is given to a defence contractor who sells it back to the military.
If you have a good idea that may have military applications, don't patent it in the UK. Instead go to the EU or US patent offices. That way it becomes impossible for the gov to steal it and give it to thier cronies.
"If you have a good idea that may have military applications, don't patent it in the UK. Instead go to the EU or US patent offices. That way it becomes impossible for the gov to steal it and give it to thier cronies." ....... Jacqui Posted Thursday 10th March 2011 15:40 GMT
The Russia/China/Japan/Pakistan route is another option. Oh, and there's always India and Israel too. In fact, there is an embarrassment of rich pickings for anyone with something which has no need for defense because it is invisibly stealthy in attack. And we haven't even started considering the number of passionate and/or crazy non-state actors out there, with more wealth than they know what to do with, and the will to change the world.
The secret in such dealings appears to be a variation of the nuclear theme ......... sell them the technology for an arm and a leg but ensure that they don't have the triggers ...... but it may be necessary to set of a big bang somewhere politically/financially sensitive just to assure every man and his dog that anything new is a viable global operating device which they need to have remote control of.
TRN, on the web now, I think this might be the same thing.
http://www.trnmag.com/
There was a lot of damn near steampunk tech (LASERpunk) in the 60s most of which went obsolete faster than it could be declassified.
..A mechanical translator using LASERS.
..Mechanical OCR using LASERS.
..Mechanical 3D radar displays; this one did not use LASERS, surprise.
and so on and so on.
Then some stuff that I worked on that is probably still classified.
I can't Google it to be sure without the black (steam or LASER powered) gyro-copters showing up.
Now of course we just emulate everything and have become cyberpunk instead.
It's interesting that you interpret the fact that the government doesn't want the information in public hands as being a sign that it is used for something spooky.
could it not just be that they don't want something with defence implications, as you say it might be useful for subs, falling into the hands of other countries that have subs?
I agree. When the article mentioned transmission ‘through steel’ , I assumed it meant at a distance. Having to have half the kit on the inside and both halves attached directly to the thick stuff isn’t really that ground-breaking, surely? Just because it is/was secret doesn’t make it clever...
With modern modulation techniques (Google OFDM) which use many discrete states (not just on/off keying) you can convey data at high rates even with only modest frequency carriers.
Your humble dial-up modem managed around 40kbits/sec over a line with sub-4kHz audio bandwidth for example. ADSL will give you 8-20Mbits/sec within a "radio" spectrum from 8kHz to 1-2MHz.
I know of cheap ($1) high-power (few 10's watts) piezo transducers which operate at around 1.7MHz... no doubt many others exist.
12Mbit/s does not mandate a channel bandwidth of 12MHz. I would guess that the signalling rate is much lower but uses complex modulation formats to achieve many bits/Hz, much like the ADSL (broadband) comms most of us use everyday. These can achieve 10-20Mbit/s in a few hundred kHz of channel bandwidth. Note I said can ;)
The Brit: Oh that's a heinously difficult problem. How can we possibly transmit data through steel ?? Maybe we should ask the BOFFINs in Cambridge. X-Rays ? Hidden dimensions ?? This is a tough problem.
The American: We need to maximize transmission capacity. Let's call Xilinx to order 500 of the fattest FPGAs and Intel for 7000 x86 CPUs we will stuff into the submarine. Then Admiral McFart and Admiral McWarmonger can do a video conference over the transducer. Proceeds to work long hours in the lab.
The German: Shannon's Channel capacity determines maximum bandwidth if you wanna know, Captain. How much bandwidth do we need if officers can use a keybord ? Ah, we need 2500 baud AND we can use some nifty crypto which <CLASSIFIED WORD A> <CLASSIFIED WORD B> .
A single 68000 and a run-of-the mill transducer from the grabbag do. Let's wire it up the next two days and then go for Beer and Schweinshaxen. Proesterchen !
In a British patent you have to declare if you think it has military implications - in which case you get the patent but it gets classified rather than published. This was part of the row about the privatization RSRE/QuinitiQ - the buyers got 50years of every secret patent for the $40M
It gets tricky for an international team, the UK govt say it has to be revealed to them if there was a Brit on the team, the US say it goes to them if it was invented on US soil (or other way around)
Back when I used to do 3letter agency stuff I asked what happened if we invented it on a conference call? Withering look from legal people - just don't right !
Transatlantic fibre optic cables - but having to have a transmitter on one end and a receiver on the other doesn't make them clever ?
The signal processing to put that much bandwidth into and out of the steel with all the thermal noise, reflections, surface scatter, multipath effect from discontinuities / grain boundaries in the metal is clever.
So let me see if I get this right.
All I have to do to to access all this super secret information on the inside of a Tempest screened room, is get inside the super secret room, plant an ultrasonic transducer and then plant another ultrasonic transducer on the outside of the room and hey presto.
Simple really.
Well apart from the 'getting into the secret room' bit. Also the usual construction of Tempest rooms is a inner and outer metal skin swandwiched between non-metal structure. Ultrasonics will not go through that too easy.
Apart from those problems........
All you need is an air gap between the inside and outside surfaces and this will never work. Ultrasonic waves can not travel in a gas but only in a liquid or a solid. Also some solids have such a high attenuation that it is impossible to get ultrasound though them, some stainless steels and nickel based alloys spring straight to mind.
Surely the easiest solution is just build your faraday cage or whatever in two layers with an air gap between them. Obviously you'd need to have a few supports to keep everything structural, but make those of a material very different to steel (and one very poor at conducting ultrasound) and you're sorted?
Coherent gamma ray emitter based on solid state X-ray generator and nuclear isomer.
intermediate energy gammas go through steel quite effectively and the whole device can be built less than a cubic centimetre powered by thermomelectric generators harnessing ambient heat differences.
The lack of macroscopic quantities of the isomer are a non issue as for the application in mind only milligram quantities in the form of a thin layer over the emitting window of the souped up Cool-X using two opposed crystals to achieve the required high keV and you're all set.
The cyclic nature is not a problem as the data can be sent in bursts.
AC, because this is slightly restricted tech...
Cordwainer Smith shows us the way to thwart spooks listening to the innards of your PC, lads. His idea of hyperspace contained something that "ate life". To protect the crew and passengers of a spaceship, the vessel was double walled, with a layer of oysters sandwiched in the cavity.
So:
I shall surround myself with all he moudering obsolete computers that are no longer fit for purpose here and power them up. Since most of them are jammed in endless boot-crash-boot cycles, all the nosy spooks will get for their clever Hi-Tech sniffing is a bunch of Microsoft fail messages, some BIOS output and hopefully a good dose of Iheartu.
They shall not surveil me as I post to El Reg!
Some JPL landers and probes operate in *very* dusty environments while others are just flat out hostile (Venus having roughly the conditions inside a sulphuric acid reactor) so they are quite keen on putting the tricky stuff inside a pressure hull with a bunch of "robust" but perhaps not too sensitive sensor heads outside
There are various PDF's on the technology around ultrasound, power ultrasound and coupling through walls.
Still pretty clever.
"a properly-equipped van parked outside a building can snoop into electronics inside even if they make no use of wireless connections. This sort of thing is expensive and very difficult – not something that most organisations have to worry about – but serious spooks can and do carry out such operations."
Sounds like the wonderful "advanced" technology the bbc claims to use to detect license fee evaders, yet according to this, such technology is difficult to use and very expensive. So either the BBC is talking shite about its tv detector vans, of it's paying a disproportionate amount of money for them and getting very little in return. Given that they aren't paying very much money at all on "detection", then I suspect they're talking shite. Again.
Don't contain any equipment, at least not any more. They are just FUD.
CRT-type displays do give off a lot of elecromagnetic emissions. It was proved that it was possible to read text on a computer monitor by picking up said emissions and decoding them. That was the basis for TEMPEST being developed to protect military command rooms from such snooping.
It is unclear whether TV detector vans really ever used this tech in Britain. There would be many difficulties proving which set was watching which channel at any given time, particularly in areas of heavy population density. There would also be plenty of interference from common household appliances such as microwaves, fridges and washing machines.
Of course, most TVs and monitors are not of the CRT type any more. LCD screens do not use high-voltage, high-frequency electronics to produce the picture, and so are much less susceptible to the technique.
Apparently there are still one or two 'detector' vans in use, but we can be reasonably certain they don't contain any actual kit capable of 'detecting' anything. TVL just rely on knocking on doors and asking persistent questions - of course, they won't confirm whether or not they have the kit as they prefer people to worry. There was an FOI request recently IIRC, which they successfully side-stepped by using crime-related reasoning.
IIRC TV detector vans looked for the local carrier wave that had to be generated by the TV, as the TV signal was sent without carrier (Single-Sideband-supressed carrier if my ailing memory is correct). That was how they could tell what channel you were watching - each channel had it's own carrier frequency. They also did handheld detectors which was how they homed in on individual flats in a block.
However, as people have commented, it was all hideously expensive, so in practice, rarely deplyed ... the hope being that a van with a spinny aerial would shock/shame the dodgers into paying.
The properly-equipped van, or at least the technology to go in it does exist. I'm less sure about the TV detector vans, but electronic eavesdropping on spurious signals is a well-established field. No doubt it has gotten more difficult with improvements in electronics and emissions standards for all electronics on the market, but I don't doubt it can still be done to a degree.
As soon as I saw that physical attachment was required, I (as did many of you) immediately realized sound waves.
An ancient trick is to listen to steel rails for an approaching locomotive. It doesn't work as well as the movies suggest because rails sometimes have air gaps (expansion gaps) every so often, but within a segment it works well enough.
It works because steel is extremely elastic and conserves energy, also, the velocity of sound in steel is very high -- 6,000 meters per second (about 340 meters/sec in air).
Strangely, a Faraday Cage, designed to block transverse electromagnetic waves doesn't stop longitudinal compression waves of ultrasound. Who'd've thunk?
Ultrasound does travel in gas, just not very well
It travels better in liquids and better still in solids (issues around elasticity not withstanding)
It is the interface between materials that is the biggest issue, where most losses occur. A coupling agent generally required (hence icon).
Nyquist applies to A-D conversion. Ultrasound (US) is not inherently analogue.
I use US professionally.
And it is fabulous at assessing steel / welds / metal fatigue / submarines....
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
