Hi all! I'm one of the co-authors. Honestly it's a dream to end up on HN with my research. As mentioned in the video we made, it has been a long road (6-7 years) to achieve this absolute moonshot of a project. I think we'll look back on this demonstration as the first experiment that truly made a distributed and real-world deployed quantum network. Not only did we use a (quantum) hardware platform capable of quantum processing, we also generated the entanglement in a way that it can be used in further quantum computations. In order for all this to work on a distributed network, we had to fully design and build the architecture to support that, both hard- and software. And we did it successfully!
Besides hard-working PhD students, another key ingredient that our research institute QuTech facilitated, was the collaboration with expert hardware and software engineers, allowing us to quickly transform new ideas into (deployable) products. A great show of what's possible when academia mixes with professional engineering.
But of course there was enough hacking and tinkering going on that it warrants to be on HN ;)
You can reply here if you have any questions, I'll be checking throughout the day. Thanks!
How hard do you expect it would be to improve the heralded infidelity from 45% to 10%?
In figure 3 of the paper [1] the heralded infidelity of entanglement is reported to be around 45%. That's not good enough for computation, but it's less than 50% which means it makes purification to arbitrarily low infidelity possible. However, the conversion rates would be pretty brutal for such a high infidelity start (e.g. millions of physical pairs consumed per logical pair good enough for use in a fault tolerant computation e.g. a target logical infidelity of 1e-6 or 1e-9).
Layman here! I have no idea what's going on but I have many questions!
- Are the photons themselves carrying quantum information?
- Does the photon link result in entangled particles in Delft and Den Haag?
- Can these entangled particles be used for communication without the optical link?
Also, I tried looking this stuff up and ran into something about quantum "repeaters" and a plans for a whole quantum network. Is this research part of working towards that? How far are we now, and what steps are still missing? Thanks!
Edit: Looks like you guys built a multi-node quantum network 2 years ago! I will have to do some more reading.
- Yes and no. The photons emitted and sent through the fiber are entangled with their electron counterparts. So we send simultaneously a photon state (entangled with electron) from Delft, and a photon state (entangled with electron) from Den Haag. Those states interfere in the midpoint (Rijswijk), and upon measurement of one photon (photon now is absorbed/measured/gone) we know that the _electrons_ of the nodes in Delft and Den Haag are entangled.
- The above also answers this question: yes!
- No. They can be used to transfer a quantum state from one place to the other, for example, which _consumes_ the entanglement (one-time use only, per pair of entangled particles). However, still classical feedback signals need to propagate for that to happen, so we still need _a_ link, preferably optical (for speed and distance). Wiki has actually a great page on teleportation: https://en.wikipedia.org/wiki/Quantum_teleportation
I'll answer to a different question on repeaters later in another comment, so check back :) Indeed, multi-node quantum network was an awesome experiment. This takes it to the next level of being able to distribute entanglement over large distances and between quantum nodes that are self-sufficient (no sharing of hardware resources between nodes).
Amplification would absorb one photon and replace it with one or more new photons. Definitely not quantum.
Personally, I always wonder why point-to-point connections are called "networks". The information is not quantum at any node, even if there are multiple nodes in a system.
Then there's "quantum internet", which makes no sense at all. What are we going to do, run direct fiber from every computer to every other computer directly? You can't hop safely or anything. Don't get me started on the total bullshit that is the "quantum repeater", now we need "quantum switch" too?
We call serial port connections things like "link", "connection", etc. We typically don't call them networks until we start linking them all together with simple routing logic that doesn't inherently require access to all the unencrypted information the packets contain and such.
To me these are all just signs that the whole scheme is/was and will forever be mostly crankery.
Quantum networking is an oxymoron. It doesn't allow end-to-end encryption and in exchange gives back extremely fragile single link security properties.
I don't think it's completely clear (to me) that quantum networking is an oxymoron. I would enthusiastically agree that its very complicated and the real world use cases are incredibly limited.
As far as your routing/switching qualms go I think they are mostly addressed by entanglement swapping? Person A and person B can each make an entangled pair and send me half, and I can (locally) do stuff which leads to the halves they keep at home becoming entangled. Then they can use teleportation or whatever to do whatever they want between themselves without me knowing anything about it.
The I can locally do stuff is completely understood theoretically/mathematically. I hand waved because this isn't a forum where those technicalities are particuarly relevant.
> What are we going to do, run direct fiber from every computer to every other computer directly?
No, you don't have to do that. A quantum network would be a web of point-to-point quantum links, with paths formed by routers choosing links. Same as a classical network.
To be a bit more concrete what an operating quantum network would look like is a bunch of routers using links to build up entanglement with their neighbors. When an endpoint wants to send a message across the network, a path from source to destination would be determined and entanglement across the links of that path would be consumed to move the message across the network [1][2]. The reason it's done this way, instead of directly sending the message, is that entanglement can be cross-checked before using it [3] and quantum networks really don't like dropping packets due to the no-cloning theorem.
> We typically don't call them networks until we start linking them all together with simple routing logic
Yeah I agree that it would be more accurate for this press release to say they made a quantum link.
> To me these are all just signs that the whole scheme is/was and will forever be mostly crankery.
Don't confuse difficulty with crankery. It'll be awhile before anyone reports an experimental realization of a true quantum network, because it'll be awhile because anyone can make a quantum router. The issue is that a quantum router is for all intents and purposes a fault tolerant quantum computer, and that is its own hard challenge being worked on separately. In particular, a quantum router needs to be able to store qubits reliably for non-trivial amounts of time, and to perform reliable operations on those qubits in order to cross-check stored entanglement.
I've worked in quantum nonlinear optics during my first postdoc 12 years ago, and back then we could only dream of the efficiency of frequency conversions that are used here. So many advances in just a decade, and most of them don't even make the news.
All those incremental changes is what made my research work indeed. As we described in the paper, the margin we had on amount of signal (dependent also on the conversion efficiency!) was small, so every % of loss anywhere in this chain of photon from emission to detection mattered.
What is actually the usecase for "quantum internet"?
Like at most i hear about quantum key distribution, but quite frankly the classical equivalents to that are just as good if not better, so what is the actual benefit?
(1) distributed computation. If you can network two quantum computers, you essentially have one quantum computer with twice the storage. Quantum networks avoid the need to build one enormous quantum computer.
(2) easier experiments. Currently, doing a loophole free Bell inequality test is hard enough that people get PhDs for it. With a quantum network that experiment is way easier, because the network solves the hard part (distributing the entanglement). You could probably also use quantum networks for other experimental tasks, like coherently linking telescopes on separate continents, though the bandwidth and computational requirements for that would probably be a bit insane.
There are also some more out there ideas, like if stock markets contain Bell inequalities then you could use a quantum network to build up entanglement that is then consumed to win those games more often which equals $$$. But it's hard to imagine concrete scenarios that would create such an inequality, nevermind one where the expected dollars gained from the quantum strategy exceeded the cost of operating the network.
A quantum internet is absolutely necessary for creating a useful quantum computer, the same way the internet (LAN) is needed to create a supercomputer. A supercomputer is essentially many computers connected together. A quantum computer that solves problems we care about will be similar: https://arxiv.org/abs/2212.10609.
As I understand, quantum key distribution cannot be beaten by classical equivalents and they're only good or better because our current quantum computers kinda suck. So the major use case at the moment is proving the tech and developing the infrastructure. The "killer app" of the quantum internet in my mind is as simple as just sending qbits around. Currently every network call involves an observation that collapses the system wavefunction. If you're looking to actually network quantum devices (say, to run distributed quantum computations) then you need quantum infrastructure.
I don't know about use case but in various distributed computing models there are problems that are provably easier for quantum computers. Unlike the classical setting where the best we have is factoring where we don't know an efficient deterministic algorithm and various problems which experimentally seem to be faster for QC (and those results often don't last long as we get better at simulating quantum algorithms classically)
I'm curious too! I'd immediately understand if it allows for speed of light communication wireless, but this is clearly wired, requiring more precision engineering than usual fibre.
I guess, but benefits should be more theoretical. Like i don't think building one will give any insight into ideas for protocols. We already understand how it would work in theory and have for a long time.
Safer mechanisms of distributing and establishing "root" keys for identify verification (so you can then use them easier with normal D-H on normal internet) is one use case I recall from 1990s.
But few years ago I heard of some other interesting uses where quantum properties were used to essentially enable DWDM-like virtual circuit routing with higher capacity - though I would have to look again if it went anywhere or into scrap heap of quantum BS.
- security - if we use quantum entanglement/teleportation to the extent I've read about how it works, then even if you still need a fiber optic cable connecting the two parties, the data is unreadable if you're not looking at physically the same wave/photons, meaning that man in the middle attack (like the ones with bending an optic cable to break it's internal reflection) is literally impossible. The data in the middle would not be readable without the receiving end entangled device, and the other side would immediately know about the attack, because an identical signal would not be readable either, as it's not the same signal anymore.
- I think the ultimate promise is transferring data without a physical link of any kind in-between. Connect two atoms, manipulate one, read the other - like ansibles in LeGuin/O.S.Card fiction. Instant interplanetary communication (which, I think, fucks up the idea of time too?)
The first one helps with physical attacks on the wire. Not a common issue that people worry about, since there are so many boxes in between that are easier to compromise that it's rarely a significant security increase if you know the wire is perfectly secure.
The second is just wrong. It is well known and proven that it's impossible to send information via quantum entanglement. It's true that there are some interpretations of QM where the wave function of the entangled pair collapses instantly the moment one side of the pair is measured. But there is no version of QM where manipulating one side of the pair has any effect whatsoever on the other, except for measurement collapsing the quantum superposition into a random classical state.
The best classical intuition for how entanglement works is that two entangled particles are like two gloves from a pair. If you put them in boxes and separate them, when someone opens a box and finds the left glove, they instantly find out that the other person has the right glove. The difference with quantum entanglement is simply that the universe only decides which glove is which when you open the box, before that they are both in a mix of the states. This makes statistical properties measurably different if you send many pairs of gloves and look at how many times certain things match.
But there really is nothing that you can do with a pair of entangled particles that you couldn't do with the pair of gloves.
I should note for completeness that, because of the different statistical properties, there is a way to send slightly more information using entangled pairs than you can with classical particles. I believe you can send 1.5 bits of information per particle, but I don't remember the exact number. This means that a quantum internet could have higher throughput at the same transmit power, which would have some relevance for very long distance wireless communication, such as communicating with a space probe.
People have dealt with the second one in sibling comments but I somewhat doubt the first one is true when you take into account sidechannel attacks on the encoding and decoding part of the transmission.
Yes I get through quantum magic you can theoretically tell if your secret has been intercepted in the quantum state because it would cause a wave form collapse but the wave form wouldn't collapse if they were listening in to your quantum computer squeaking and buzzing and decoding those noises or timings or reading its heat signature etc, or getting your operator drunk and finding out their dog's name or partner's birthday and using it as their password, or kidnapping them and hitting them with things until they voluntarily give you their password etc. All those types of attacks would still work and still be just as undetectable as they are in classical encryption. ie all the most effective forms of attack are still just as effective in a quantum case.
I think it's a very interesting area of research but this whole idea of uncrackable codes is a stretch.
As far as In understand it (not very much) you can listen in on the transmitted keys, but the interaction can be statistically(!) measured and suspicious bits can me omitted (the wiki is quite comprehensible: https://en.wikipedia.org/wiki/Quantum_key_distribution?wprov...).
There are different protocols, some more and some less quantum and most rely on classical, encrypted channels and trusted nodes in addition to the quantum channels.
One thing is for sure: you can’t send information faster than light with this or any other kind of quantum communication as two entangled qubits are basically two RNGs that are correlated. You’d just get noise without an additional classical, not FTL, data link (please, somebody with expertise: help!)
As far as I know, they still need classical encryption methods (with something like shared secret key or public key for authentication) to detect active man in the middle attacks where the attacker prevents the parties connecting to each other and then pretending to both parties to be the other party by creating his own "messages" as if they came from the other party. Or at least to have some kind of additional trusted physical medium where it is impossible to prevent the parties communicating directly, capturing their "messages" and then sending your own modified "messages" instead -- perhaps based on some kind of timing etc.
And if you still have to rely to classical encryption methods to make sure you know the identity of the other party (to prevent active man in the middle attack), why not just use classical encryption methods for everything else as well, instead of using quantum key distribution?
You don't need "classical encryption" for quantum key distribution. With QKD you can provably detect if a MITM attack happened. With classical methods you can never be 100% sure, although how much of that matters in practice is another question.
> You don't need "classical encryption" for quantum key distribution. With QKD you can provably detect if a MITM attack happened.
This is incorrect. QKD can detect passive mitm only. It cannot detect an active mitm.
Which is the main reason its overhyped, since as cool as QKD is, you still need active mitm prevention, so you have to rely on classical crypto anyways.
To disperse some of the hype here around using this for "uncrackable" key exchange: QKD has been a product of choice for cybersecurity conmen for decades.
Besides hard-working PhD students, another key ingredient that our research institute QuTech facilitated, was the collaboration with expert hardware and software engineers, allowing us to quickly transform new ideas into (deployable) products. A great show of what's possible when academia mixes with professional engineering. But of course there was enough hacking and tinkering going on that it warrants to be on HN ;)
You can reply here if you have any questions, I'll be checking throughout the day. Thanks!
reply