Slashdot Mirror
Securing Fiber Using Light Polarization
screenbert writes: "A new and novel way of
communicating over fiber optics is being developed by physicists supported by
the Office of Naval Research. Rather than
using the
amplitude and
frequency
of
electromagnetic waves, they're using the polarization of the wave to carry
the signal. Such a method offers a novel and elegant method of secure
communication over fiber optic lines. This
press release has more information. Of course I always thought that fiber
was always pretty secure anyway since it's a lot harder to tap than copper."
Quite likely nobody knows which fiber is the one you need. As I understand it, after a break and splice, the fibers may be scrambled -- so the receivers on either end go through a discovery process to figure out how things are hooked up now.
However, tapping a fiber isn't that difficult in principle -- you bend it just enough so some light escapes.
Even if you have the equipment, you have to figure out which fiber out of 288 or more is the one you need, and the documentation is usually kept locked up tight.
Which is just security through exclusivity/obscurity, right?
"I'm not impatient. I just hate waiting." - My Dad
This method is neither new or novel, it's called Quantum Encruption. You can read a quick primer Here. By using polarized photos, you can trasmit bits that will be impossible to intercept without being detected. Research labs have been working on relaible, long-distance implementation for years.
As for knowing what fiber goes where, again, good luck getting the info. I worked for a fiber optic mapping company for some time (hence why I'm posting AC), I've seen some of the maps and info the companies have for their -own- networks. Many companies are in the process of digitizing their maps, but most often the ones they have now are paper, fairly cryptic, with only one/two people really being 'in the know' as to what they mean, per. region.
Thats the theory behind quantum encryption, in which single photons are used to create a shared key by playing tricks with polarisation. The important point is the words "single photon".
However QE cannot work over long distances because photons get lost (i.e. attenuation). General purpose signalling sends a lot of photons so that at least a few get through (I think the detection level for general purpose detectors without special cooling is around 70 photons). They also get amplified. I'm not sure if fibre amplifiers maintain polarisation. If not then this technique is just an interesting novelty.
So tapping would be easy. Just put the signal through a splitter (e.g. a bend in the fibre) and route your half of the signal to a decoder that works in the same way as the official one. The other end sees a 3dB drop in signal, but thats probably too small to be noticed.
Where this might be important is increased bandwidth. At the moment fibre transmission uses binary keying: send photons for 1, no photons for 0. Polarisation modulation means that you could use several different angles, and hence encode more than one bit per light pulse.
But don't get too excited about the bandwidth either. The limiting factor on bandwidth at the moment is the routers at the end of the fibre. We can pump terabits down a fibre in the lab, and 100 Gbit is pretty straighforward to do in the field. But put ten 100Gbit links into a router and you have to have a machine that can switch 1 Tbit. If the average packet is 1.5kbytes (Ethernet frame) then thats around 83 million packets per second. Even with hardware assist thats an awful lot of address table lookups per second.
Paul.
You are lost in a twisty maze of little standards, all different.
it'll be secure until anyone with the right
dosh can get their hands on one of these
recievers
Not quite, compare the Enigma encryption machine from WWII. The machine wasn't the encryption, just the device (although the machine itself was quite clever). Without knowing the proper setting for the machine, it was near worthless. The allies had their hands on Enigma for several years before they came up with a cryptanalytic method (kudos to the Poles!) that made the physical machine they had worth something. The encryption is in the signal and noise, not in the machine that reads it.
I have a question on that anyway.
I know you can't put a quantum crytography signal through an EDFA, thus making lots of copies of the signal photons and giving you enough chances to beamsplit and measure the polarisation states.
How do I know this? Because it wouldn't be a good system if you could. What I want to know is why doesn't this work? What fundamentally stops this happening?
It can't be that Eve might NOT split off all but one of the amplified bunch of photons for any individual bit and thus giving the game away to Alice because Eve could just retransmit from scratch.
Is it instead something about the beamsplitting process? I seem to remember a presentation at Uni from one of the theory guys which implied that 2 identical photons (such as the original and a copy out of an amplifier) are not independantly beamsplit but that instead take to reflection or transmission output path from the beamsplitter as a pair.
Is that right? Or is it something else entirely.
If there is anyone who can reply, that would be great. I know all the experimental side of these things, I built Erbium fibre ring lasers and looked at their output polarisation states for my PhD. I just don't have the quantum theory knowledge.
This is not. This is merely using a polarity of the lambda and a timing seed. The reason it's "secure" is because you would have to have the hardware and the timing key to strip out the intelligence. Right now you would have to put a filter in front of the beam to do this. For the time being, it would be readily obvious if the FO were tampered with by the signal loss. On the other hand, if you know the capacitive and inductive reactants of your phone line you could tell if someone was tapping it between you and the SAC box. So now they wire tap at the CO or before the SAC box at least. I don't think it would take too much work to come up with a way to read the intelligence in a non-intrusive way. Not so long ago plane Jane FO was thought to be secure for the same reason. Now there are OTDR's that can read the signal through the cladding. This rises a question, I wonder how they plan on using repeaters for signal regeneration? What IS cool about this, is the fact you could use this to cram even more intelligence on a single lambda in much the same way QAM allows for higher data rates with a more efficient use of frequency. I know that Bell Labs played with this while working on their optical routers. But hey they did bring us QAM and FO. Now if only they could use that brain power to figure out a way to make a profit! I see a lot of post tossing around the "Quantum Entanglement" phrase. This is not Quantum entanglement. ( NOTE this is a gross over simplification but it makes the point with out the math) In a nutshell Quantum entanglement works like this: Quantum entanglement is when two particles become Quantumly entangled or one particle is split in a way that produces two Quantumly entangled partials. The particles will have an inverse relationship or an "inverse spin". Imagine one spins left the other right. If I keep the left spinning particle and watch it while I transport the right spinning particle to where I want to communicate with. I can effect change on the right spinning partial by affecting my left spinning particle.. Like wise if someone were to "intercept" the right spinning partial I could watch my left spinning particle be affected by the right spinning particle. In this way I could have a clue my data had been tampered with or read. The Problem with this is that to the best of my knowledge no one have ever managed to keep Quantum entanglement for more the short distances. Then math says we should be able to but no one has done it yet. However, I haven't been reading up for a few years so I might have missed it. It's 0300 I've been drinking and I have an Internet connection. :-P