Quantum communications leap out of the lab

14

China begins work on super-secure network as ‘real-world’ trial successfully sends quantum keys and data.

Cybersecurity is a step closer to the dream of sending data securely over long distances using quantum physics — spurred by two developments.

This week, China will start installing the world’s longest quantum-communications network, which includes a 2,000-kilometre link between Beijing and Shanghai. And a study jointly announced this week by the companies Toshiba, BT and ADVA, with the UK National Physical Laboratory in Teddington, reports “encouraging” results from a network field trial, suggesting that quantum communications could be feasible on existing fibre-optic infrastructure.

Conventional data-encryption systems rely on the exchange of a secret ‘key’ — in binary 0s and 1s — to encrypt and decrypt information. But the security of such a communication channel can be undermined if a hacker ‘eavesdrops’ on this key during transmission. Quantum communications use a technology called quantum key distribution (QKD), which harnesses the subatomic properties of photons to “remove this weakest link of the current system”, says Grégoire Ribordy, co-founder and chief executive of ID Quantique, a quantum-cryptography company in Geneva, Switzerland.

The method allows a user to send a pulse of photons that are placed in specific quantum states that characterize the cryptographic key. If anyone tries to intercept the key, the act of eavesdropping intrinsically alters its quantum state — alerting users to a security breach. Both the US$100-million Chinese initiative and the system tested in the latest study use QKD.

The Chinese network “will not only provide the highest level of protection for government and financial data, but provide a test bed for quantum theories and new technologies”, says Jian-Wei Pan, a quantum physicist at the University of Science and Technology of China in Hefei, who is leading the Chinese project.

Pan hopes to test such ideas using the network, along with a quantum satellite that his team plans to launch next year (see Nature 492, 22–25; 2012). Together, he says, the technologies could perform further tests of fundamental quantum theories over large scales (around 2,000 kilometres), such as quantum non-locality, in which changing the quantum state of one particle can influence the state of another even if they are far apart, says Pan.

Sending single photons over long distances is one of the greatest problems in QKD because they tend to get absorbed by optical fibres, making the keys tricky to detect on the receiver’s end.

This is “a big challenge for conventional detectors”, says Hoi-Kwong Lo, a quantum physicist at the University of Toronto in Canada. But technological breakthroughs in recent years have significantly reduced the noise level of detectors while increasing their efficiency in detecting photons from just 15% to 50%.

Vast improvements have also been made in the rate at which detectors can ‘count’ photon pulses — crucial in determining the rate at which quantum keys can be sent, and thus the speed of the network. Counting rates have been raised 1,000-fold, to about 2 gigahertz, says Lo.

The breakthroughs are pushing the distance over which quantum signals can be sent. Trials using ‘dark fibres’ — optical fibres laid down by telecommunications companies but lying unused — have sent quantum signals up to 100 kilometres, says Don Hayford, a researcher at Battelle, a technology-development company headquartered in Columbus, Ohio.

Written By: Jane Qiu
continue to source article at nature.com

14 COMMENTS

    • In reply to #1 by Stafford Gordon:

      I wonder how long it’ll be before a freelance geek cracks this!

      If I’m understanding the research correctly it will be a very long time, possibly infinite. There already are security systems, e.g. public key encryption in place that no one but groups like the NSA can crack and even they require a shit load of computing power or back door access to crack them.

      Of course no security is fool proof because there are always humans involved. You may have a security system that is virtually hack proof but if you write your password on a Postit and leave it on your screen…

      • In reply to #2 by Red Dog:

        In reply to #1 by Stafford Gordon:

        I wonder how long it’ll be before a freelance geek cracks this!

        If I’m understanding the research correctly it will be a very long time, possibly infinite. There already are security systems, e.g. public key encryption in place that no one but groups like the NSA…

        Thanks for that; but I’m so far behind the curve on this subject that I think it’s best if I just stick to gardening!.

  1. Could lasers be used to transmit high speed data for photon pulse codes as opposed to fibre optics….?
    The recent article on astronauts using lasers to beam video streams back to earth fast – sounded like a great leap forward in technology…maybe not the same applications though !

    • In reply to #3 by Light Wave:

      Could lasers be used to transmit high speed data for photon pulse codes as opposed to fibre optics….?
      The recent article on astronauts using lasers to beam video streams back to earth fast – sounded like a great leap forward in technology…maybe not the same applications though !

      We already have a form of this, and have had it for a long time. It uses infrared light, and you can use it to transmit data between phones, or between a phone and a computer. It’s secure because the beam can’t be spied on, but it has some really severe limitations in terms of distance. Look up “IrDA” in Wikipedia for some information.

  2. There is something about this story which does not ring true for me.

    The basics are:

    Plain text … Encryption … Cypher text A … Transmission … Cypher text B … Decryption … Plain text

    Next is the use of the word “photon”. Photons and electrons are both fundamental particles in the partial physicists’ Standard Model. On that basis, how does the use of photons change my above environment? They are talking about a fibre-optic network, but no-one is saying how photon transmission is different to electron transmission. Are we being asked to believe that fibre networks, unlike traditional electron networks, transmit such waves end-to-end unchanged? It seems to me that, wherever signal amplification takes place (e.g. In an erbium-doped optical amplifier), photons are subjected to similar transmission noise problems associated with electrical transmission.

    I hasten to add that I am not an expert and I may have forgotten a few things as I haven’t worked in telecoms for over a decade. Can anyone out there educate me … please?

    Even if I am wrong about the above, which seems likely, I don’t understand this report for an entirely different reason. Even if I can create, within a photon, a state that can only be read once without corruption this does not put such communications beyond cryptanalysis. Just as electronic-binary can be cracked, so too can photon-tertiary. My best bet would be on copying the transmission stream. How do we do that without subjecting the original transmission to a read-corrupt-detect alert? Quantum computing.

    Peace.

    • In reply to #5 by Stephen of Wimbledon:

      There is something about this story which does not ring true for me.

      The basics are:

      Plain text … Encryption … Cypher text A … Transmission … Cypher text B … Decryption … Plain text

      Next is the use of the word “photon”. Photons and electrons are both fundamental particles in the part…

      The important thing is that the transmission uses single photons and cares about the actual polarisation state of each photon. Normal transition just sends a whole bunch of photons (or electrons if on wire) and cares about how much energy was sent at what frequency. It does not measure the quantum state of the quanta in the data stream.

      There is a theorem in Quantum mechanics know as the “no cloning theorem”, which says that you cannot take quantum information and copy it flawlessly. Since the polarisation of a photon is a quantum mechanical property, a third party will not be able to make a copy of the data stream without destroying the original or getting a flawed copy.

      Of course, the actual process is much more complex than this.

    • In reply to #5 by Stephen of Wimbledon:

      There is something about this story which does not ring true for me.

      AIUI, the photons have a pair of indeterminate properties (both kinds of polarisation, i.e. the orientation of the electromagnetic oscillations of the photon, I think) similar to the up or down state of an electron. It is, apparently, possible to send unmeasured photons down an optical fibre. One pre-shares a sequence of which one of these states to measure with the recipient as one would share a normal cryptographic key. The trick seems to be that it is only possible to measure one of the states. If someone is trying to intercept the data, they don’t know which to measure, and lose one of the states. If they try to send the message on, the recipient can use ordinary error correction (e.g. check sums) to determine that some of the data has been randomised, and know that tampering has occurred.

      The thing is, the same would be true if there was a signal problem. I guess there would have to be a retransmit part of the protocol. The question for me is how signal strength is retained. Do they just use a very, very bright source, i.e. lots of identically encoded photons? If there’s some way to amplify the signal, doesn’t that mean an interceptor could amplify and divert part of it? Isn’t that also true if there are a lot of them? I guess one could check for relative dimness in the latter case.

    • In reply to #5 by Stephen of Wimbledon:

      Even if I am wrong about the above, which seems likely, I don’t understand this report for an entirely different reason. Even if I can create, within a photon, a state that can only be read once without corruption this does not put such communications beyond cryptanalysis. Just as electronic-binary can be cracked, so too can photon-tertiary.

      My understanding of the article was that it was a very specialized application. BTW, I didn’t really follow it completely either so I could be wrong. But it seemed to imply that the purpose of this was only to communicate things like passwords and keys. I.e., the actual information isn’t that critical, but knowing if someone else knows the information is the critical thing.

      So that was the benefit, due to the quantum technology it’s theoretically impossible (I think) for someone else to access the information without the parties that are supposed to have access knowing that someone else accessed it so it provides iron clad security. If you know a password may have been intercepted you don’t use it and you just create and resend another one.

  3. I don’t quite understand…
    Quote “Conventional data-encryption systems rely on the exchange of a secret ‘key’ — in binary 0s and 1s — to encrypt and decrypt information”. But my understanding of public key encryption is that the secret key (aka private key) is NOT exchanged – only the public key (useless to a hacker) is exchanged. In other words, one key is used to encrypt and a DIFFERENT key (not exchanged) is used to decrypt. Which is why Public Key Encryption is so successful – a hacker can intercept your entire message without any chance of decrypting it. Why is this proposed quantum system any different?

    • In reply to #11 by hairybreeks:

      I don’t quite understand…
      Quote “Conventional data-encryption systems rely on the exchange of a secret ‘key’ — in binary 0s and 1s — to encrypt and decrypt information”. But my understanding of public key encryption is that the secret key (aka private key) is NOT exchanged – only the public key (…

      That puzzled me as well. I’m not sure but I think this may be referring to symmetric encryption algorithms, i.e. not public key which are asymmetric, in symmetric encryption the users have to initially share a private key. From the wikipedia article I linked to earlier on this thread:

      Public key algorithms, unlike symmetric key algorithms, do not require a secure initial exchange of one (or more) secret keys between the parties.

  4. Many thanks to:

    Stuart Coyle

    PERSON

    Red Dog

    and Urn

    The two companies selling this gear commercially are obviously in the business of exploring for cracks in their systems. As the latest tech quantum encryption has the advantage that people need to study it before they can crack it.

    Peace.

Leave a Reply