Researchers Break World Record For Quantum-Encrypted Communications (engadget.com) 53
Researchers in Beijing have set a new quantum secure direct communication (QSDC) world record of 102.2 km (64 miles), smashing the previous mark of 18 km (11 miles), The Eurasian Times reported. Engadget reports: Transmission speeds were extremely slow at 0.54 bits per second, but still good enough for text message and phone call encryption over a distance of 30 km (19 miles), wrote research lead Long Guilu in Nature. The work could eventually lead to hack-proof communication, as any eavesdropping attempt on a quantum line can be instantly detected. QSDC uses the principal of entanglement to secure networks. Quantum physics dictates that entangled particles are linked, so that if you change the property of one by measuring it, the other will instantly change, too -- effectively making hacking impossible. In theory, the particles stay linked even if they're light-years apart, so such systems should work over great distances.
The same research team set the previous fiber record, and devised a "novel design of physical system with a new protocol" to achieve the longer distance. They simplified it by eliminating the "complicated active compensation subsystem" used in the previous model. "This enables an ultra-low quantum bit error rate (QBER) and the long-term stability against environmental noises." As a result, the system can withstand much more so-called channel loss that makes it impossible to decode encrypted messages. That in turn allowed them to extend the fiber from 28.3km to the record 102.2 km distance. "The experiment shows that intercity quantum secure direct communication through the fiber is feasible with present-day technology," the team wrote in Nature.
The same research team set the previous fiber record, and devised a "novel design of physical system with a new protocol" to achieve the longer distance. They simplified it by eliminating the "complicated active compensation subsystem" used in the previous model. "This enables an ultra-low quantum bit error rate (QBER) and the long-term stability against environmental noises." As a result, the system can withstand much more so-called channel loss that makes it impossible to decode encrypted messages. That in turn allowed them to extend the fiber from 28.3km to the record 102.2 km distance. "The experiment shows that intercity quantum secure direct communication through the fiber is feasible with present-day technology," the team wrote in Nature.
Entanglement? (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:3)
When you're communicating, your entangled particles don't need to be "connected" by anything. The very act of interfering with the distribution of the entangled particles is detectable. The very act of observing the particles has certain mathematical properties that permits those who are observing them 100% confidence that they are entangled and nobody else has a "copy" to observe/spoof/MiTM.
Re: (Score:2)
When you're communicating, your entangled particles don't need to be "connected" by anything. The very act of interfering with the distribution of the entangled particles is detectable. The very act of observing the particles has certain mathematical properties that permits those who are observing them 100% confidence that they are entangled and nobody else has a "copy" to observe/spoof/MiTM.
The point of quantum is forward secrecy not confidentiality.
100% confidence that nobody else has something is worthless when you don't know who you are communicating with. Authentication in QSDC/QKD is supplied classically by keeping secrets. Without it the system becomes a paperweight no more useful than anonymous DH.
Intentional sabotage (Score:2, Interesting)
What is stopping a malicious entity from intentionally eavesdropping just to poison the entire line?
Imagine connecting multiple silent devices. When one is found and detected then another turns one.
This would make the entire "secure" line unusable.
Re: (Score:1)
i have to think that the same old modulation techniques always used, that and multiplexing will go a long way to help secure. i'm assuming a very high frequency line signal....
Re: (Score:2)
Re:Intentional sabotage (Score:4)
That's the same as injecting noise into existing non-quantum communications wire. They can isolate where you are disrupting the signal and award you a cruise missile really fast.
Re: (Score:1)
Oh I always wanted a cruise missile! Know what the second hand price is at these days?
Re: (Score:1)
Re: (Score:2)
Nothing. It's intended to maintain privacy at any cost, including total disruption.
Re: (Score:2)
Nothing. Communication secured with quantum entanglement simply guarantees you can tell if somebody has intercepted the data. It's secure from eavesdropping, not secure from disruption.
What's the market for this? (Score:3)
Do people seriously plan to optical switch entire large scale networks? Otherwise what's the point of this vs. OTP pools that are cheap and can effectively last a very long time?
If you are that paranoid and your secrets are so important why would you rely on anything other than XOR?
Re: (Score:2)
The problem with OTP is that you need to supply the same amount of OTP as the data you are sending, because if you re-use any of it your system is compromised.
This is more useful for backbones and specialist links, so packet switching on large scale networks isn't a consideration. Intercontinental cables, satellite to ground, inter-bank links, that kind of thing.
Re: (Score:3)
The problem with OTP is that you need to supply the same amount of OTP as the data you are sending, because if you re-use any of it your system is compromised.
OTP is an analogue of quantum channel and can do everything it can do only better. (Trusted, no quantum infrastructure)
It can be used to provide forward secrecy or key any number of ciphers which are inherently quantum safe. There is no requirement for a 1:1 correspondence between data and OTP. Given bitrates mentioned in TFA even if there were 1:1 an OTP pool would last forever.
This is more useful for backbones and specialist links, so packet switching on large scale networks isn't a consideration.
This is why I don't understand the value proposition. It's one thing to use this scheme to provide forward secrecy for a large
Re: (Score:2)
The other issue with OTP is that it needs to be agreed before hand. If someone gets your OPT pool, you are completely compromised, for the entire period that the pool is in use.
With quantum key distribution the key can be generated just prior to use and discarded when finished, which greatly reduces the risk of it being compromised, and limits the time period affected by a compromise.
Re: (Score:2)
The other issue with OTP is that it needs to be agreed before hand. If someone gets your OPT pool, you are completely compromised, for the entire period that the pool is in use.
This is true of literally everything.. QSDC, QKD or speaking in code over a string can...everything. There must always be prearrangement of keying materials and guarded secrets in order to authenticate the channel.
If someone gets your guarded secret any future communications are completely compromised regardless of whether you are using quantum channels or OTP. Compromise of guarded secrets is always fatal.
Prior communications are safeguarded from future compromise simply by irreversible erasure of OTP as
Re: (Score:2)
Do people seriously plan to optical switch entire large scale networks?
Are you seriously asking that of China? The place where thousands of km of high speed rails was laid in mere two decades, ultra-high voltage transmission lines were laid across the length of the country from west to east, and a newly made vast water pumping system moving water from south to north. Not to mention huge airports, brand new autonomous ports, extensive 5G networks everywhere.
So, yeah, I would not be surprised a bit if China build a huge network backbone using quantum-encrypted links.
Re: (Score:2)
Are you seriously asking that of China? The place where thousands of km of high speed rails was laid in mere two decades, ultra-high voltage transmission lines
I'm not sure what country or rail and transmission lines have to do with the question raised? The problems associated with ditching all buffers, repeaters/amplifiers in an a real world optically switched network seem impractical unless you are content to limit the network to a small geographical area.
So, yeah, I would not be surprised a bit if China build a huge network backbone using quantum-encrypted links.
Quantum encrypted links is a very different thing than quantum encrypted peers. One can easily provide forward secrecy between large P2P links using OTP pools that effectively last forever given bitrates desc
Re: (Score:2)
The parent was just saying that being large scale, complex, and difficult isn't exactly a barrier to entry for China.
Re: (Score:2)
Do people seriously plan to optical switch entire large scale networks?
Yes. Why is that even a question. The only actual question is when. In theory you can get much better efficiency that way... eventually
Re: (Score:2)
Do people seriously plan to optical switch entire large scale networks?
Yes. Why is that even a question. The only actual question is when. In theory you can get much better efficiency that way... eventually
Optically switching an entire network with no buffers or amps/repeaters is not efficient. Quantum channel bandwidth decreasing exponentially with path length is not efficient.
Re: (Score:3)
There's a lot of impracticality here. Including using quantum entanglement at all when you could communicate faster using a set of standard resistors over copper.
Both sides start by switching in 5K (for example) resistors so that each side can characterize the resistance of the circuit. Then, each side switches in a resistor corresponding to the first symbol they want to transmit (or a random one if they only expect to recieve). Each side measures the resistance of the circuit. Since they know the resistanc
Re: (Score:3)
Re: (Score:2)
And so you use the resistors to share the key. Still no way to MitM the circuit, key still gets shared. No need for hyper expensive equipment.
Re: (Score:2)
Re: (Score:2)
If it's been screwed with, the hashes will fail and you have detected the problem.
Meanwhile, on the quantum side, how do you know you're not talking to Eve rather than Alice? EC signing, I'd imagine. But then what use is the quantum part?
Re: (Score:2)
"would tend to"
It's not like your "solution" isn't known to everybody. But it's easily MITM'd - granted, your "disruptions" would have to happen outside a window of observation, but that's exactly the point. Or if you observe the disruption, you have no confidence in knowing if or when that interception ever goes away. You'd have to throw out the line or physically inspect the entire length of it. We already have extremely secure methods of communication that are effectively what you're describing.
Quantum e
Re: (Score:2)
Quantum encryption makes MITM attacks literally impossible. The two parties at either end could be knocked unconscious, have their minds wiped, etc etc .. when they wake up and communicate via their quantum entangled line, they don't have to worry the line having been messed with. At all. They can both be 100% confident that what they share over the secure quantum line is guaranteed not to have been intercepted by dint of the laws of physics.
So how does this work? Do quantum states have little to and from headers embedded in them? Does god examine those headers and infallibly make sure that only the right person receives them?
How does either party determine which party the other is communicating when no prior basis for such knowledge is stipulated to exists?
Re: (Score:2)
https://en.wikipedia.org/wiki/... [wikipedia.org]
Whenever this topic comes up, I highly recommend https://www.amazon.ca/Quantum-... [amazon.ca]
High school (but not "simple") math that shows you the actual mathematical properties and reasoning why the probabilistic behavior of entangled particles permits you to know that nobody else has intercepted or is "spoofing" two entangled particles (aka two qubits)
You're not "sending" particles around apart from the original engagement and moving the pair of qubits away from each other physicall
Re: (Score:2)
High school (but not "simple") math that shows you the actual mathematical properties and reasoning why the probabilistic behavior of entangled particles permits you to know that nobody else has intercepted or is "spoofing" two entangled particles (aka two qubits)
This doesn't answer my question. How does either party determine which party the other is communicating when no prior basis for such knowledge is stipulated to exists?
To put it another way if you have no idea who you are talking to what is the benefit of the being assured nobody else has intercepted your message? Does cryptography have any meaning or worth without a basis in trust?
And you need to also communicate in a way that needn't be secure outside those cans. That's your "headers" in your metaphor.
With QKD alone you basically end up with the equivalent of an uncrackable anonymous DH which is rather worthless for secure
Re: (Score:2)
And btw this assumes you're confident enough that not seeing a "disruption" means it hasn't been tampered with. That's a (possible very confident) assumption, not a known fact. So you might appreciate that that may not cut the mustard for certain applications and requirements of those nation states with every resource at their disposal seeking the highest level of security possible if the alternative is "100% positive" instead of "extremely sure".
Re: (Score:2)
Or someone personally delivers a thumb drive with a billion keys on it and all you have to do is exchange offsets into the file.
Otherwise, Eve sits in the middle and separately agrees on a key with Alice and another with Bob. She then translates the subsequent communication.
Re: (Score:3)
It'll never replace sneakernet. (Score:2)
Re: (Score:1)
There's also the station wagon.
0.54 bits/s is not enough for a phone call (Score:1)
Re: (Score:2)
It's just an improvement on what has been done in the past. None of this breaks new ground in theory. Which is not to say it's not amazing and cool, but the work upon which this work extends has been known, accepted and reproduced reliably by the peer group. There isn't really a compelling reason to default to distrust here.
How does entanglement help communication? (Score:2)
Yes, I'm stupid. EVERYONE knows this already because we've all read the papers and studied the quantum maths and stuff. I'm dumb and lazy and dim witted.
Ok, now that we've got that out of the way.
How does entanglement help with communication between parties that don't already know what's being sent between them? There's really got to be some core principle here that I'm just not considering here.
I get that entangled particles change states in coordination with each other but how does that help when neith
Re: (Score:3)
How does entanglement help with communication between parties that don't already know what's being sent between them?
If you are sending top secret encrypted information and suspect that NSA has a practical attack on prime factorization problem that implements key establishment methods your choices are limited to fairly archaic methods. Having quantum entanglement channel, regardless of costs and bandwidth limitation, would allow you to sidestep such potential attack.
Re:How does entanglement help communication? (Score:4, Informative)
"How does that help without some sort of communication between the two ends coordinating the effort?"
You do need that extra line of communication, which is why quantum entangled communication does not enable superluminal communication. It only facilitates communication that you can be 100% confident was not intercepted or eavesdropped upon.
this will teach you everything you need to know about this stuff, using mostly only high school algebra
https://www.amazon.ca/Quantum-... [amazon.ca]
Re: (Score:2)
Thanks!
Re: (Score:2)
It's an insanely fascinating book. I really do recommend you read it if you're interested in this stuff, because not only is it pretty mind blowing, but it really does a great job of breaking down how it works mathematically and you come away with confidence that the way we want to think it works (that particles have hidden local values that are set before we observe them) just isn't true. And once you observe it, the probabilities collapse and as long as you continue to observe in the exact same way, you g
How slow? (Score:2)
>Transmission speeds were extremely slow at 0.54 bits per second, but still good enough for text message and phone call encryption over a distance of 30 km (19 miles)
At half a bit per second, how is that good enough for phone call encryption?