3
$\begingroup$

Quantum key distribution (QKD) promises unconditional security in data communication and is currently being deployed in commercial applications. Nonetheless, before QKD can be widely adopted, it faces a number of important challenges such as secret key rate, distance, size, cost, and practical security. nature.com article

What are challenges in quantum key distribution ?

$\endgroup$
  • $\begingroup$ Excellent link. $\endgroup$ – Paul Uszak Sep 11 '18 at 16:01
  • $\begingroup$ Actually, this isn't a question; it really is advocacy for the opinion "QKD isn't practical" (and the fact that I basically agree with this doesn't change the fact that this really isn't a question) $\endgroup$ – poncho Sep 11 '18 at 20:27
  • $\begingroup$ In light of opposition to the widely acknowledged benefits of QKD, it might be worth repeating "You can fool all the people some of the time, and some of the people all the time, but you cannot fool all the people all the time." Since we deal with probabilities here, what is the probability that Toshiba and ID Quantique are fooling all the people all the time? $\endgroup$ – Paul Uszak Sep 11 '18 at 22:24
  • 1
    $\begingroup$ @PaulUszak Demonstrably 0, proof being in the form all of the reasoned and skeptical users on this very site (as well as noteworthy cryptographers such as DJB) $\endgroup$ – Ella Rose Sep 11 '18 at 22:44
  • 2
    $\begingroup$ @PaulUszak: "In light of opposition to the widely acknowledged benefits of QKD..."; might you highlight what those benefits might be (compared to the alternatives)? If you use QKD to generate (say) AES keys, then the system is no stronger than AES; that begs the question: why don't you just use AES to distribute the keys? Such a system would be no weaker (and far cheaper/more flexible). If you say "someone with access to an AES keyed system can find the master AES key", well, someone with access to a QKD system can modify it to appear to be secure, but not actually be secure. $\endgroup$ – poncho Sep 12 '18 at 3:06
5
$\begingroup$

The article cited in the question (Eleni Diamanti, Hoi-Kwong Lo, Bing Qi, and Zhiliang Yuan: Practical challenges in quantum key distribution, in npj Quantum Information 2016, a Nature Partner Journal) gives a list of challenges facing Quantum Key Distribution, but forgets several important ones:

  • Not fulfilling the market demand for simplified key distribution. When Man in the Middle attack is a threat, QKD only extends a pre-existing initial secret key, which still must be distributed by trusted means such a courier. QKD is thus not easier to use than classical key distribution and offers no clear advantage compared to distribution of a USB key with a One Time Pad by that same courier.
  • QKD's claimed security covers keys distribution, not data transmission at commercial speed. The later still relies on classical symmetric cryptography (lacking mathematical proof), with QKD protecting only the low trickle of data that constitutes the keys for classical crypto. Therefore, use of QKD does not unconditionally prevent interception of the bulk data for later decryption by hypothetical improved attacks.
  • Incompatibility with deployed infrastructure: existing active electronics at optical fiber ends, including in repeaters (often underwater), are incompatible with QKD. Some of the existing fibers might be re-purposable, but at extremely low QKD rate, which worsens the above issue.
  • QKD does not scale to a global network. What's currently experimented is end-to-end only, or (marginally) with special optical routers that further reduce the usable distance. Blueprints for large networks require trust in the routers, when classical crypto does not.
  • Complexity of the indispensable information-theoretic (mathematically provable) paraphernalia necessary for QKD to work both reliably and safely. A most cited method for that appeared only 9 years after the birth of QKD, and despite its complexity leaves it to the practitioner to deal with power outages other than by using another courier for a fresh initial secret key.
  • Lack of security certification: there was no security-certified Quantum Key Distribution system last time I checked, nor any planned reference framework for such certification. Market and laws demand certification (rightly so: several QKD systems have been hacked).
$\endgroup$
  • $\begingroup$ Err, do you realise that classic cryptography has nothing whatsoever to do with quantum key distribution? The article and the subject of QKD is about the secure transmission of encryption keys. Period. You seem to be criticising something else entirely. $\endgroup$ – Paul Uszak Sep 11 '18 at 17:56
  • $\begingroup$ I do realize that QKD can (in practice: only) be used to distribute the keys of classical crypto, and have now made that clear. $\endgroup$ – fgrieu Sep 11 '18 at 18:58
0
$\begingroup$

Your article concludes with:-

In this review, we have discussed important challenges in practical QKD.

They are all listed therein.

Consider the providence of this article. Nature is one the most respected publications in all of science. If they cure cancer, it will probably be first published in Nature. I'm not sure what we lowly minions might be able to add, unless you are looking at a more specific issue. Although the 182 references will probably cover them all. Having been accepted 29 May 2016, it's pretty up to date too.

I suggest that Yuan's contribution adds a very important commercial aspect to the work. Toshiba are interested ultimately in profit and prestige (as is correct for a company), so it is in their interest and the article's breadth that real world QKD is covered.

$\endgroup$
  • 2
    $\begingroup$ The article is not published in Nature. It is published in a Nature Partner Journal. Further, being published in Nature is not an absolute assurance of seriousness. $\endgroup$ – fgrieu Sep 11 '18 at 16:24
  • 2
    $\begingroup$ The question was What are challenges in quantum key distribution ?. This answer appears to do nothing other than praise the journal that the article was supposedly written in (which as fgrieu notes is not even accurate) and does not address the actual question. $\endgroup$ – Ella Rose Sep 11 '18 at 16:27
  • $\begingroup$ @fgrieu Yup, missed the partnership issue :-( Although, it seems like a pretty comprehensive review and nevertheless resides under the nature.com domain. Also, it has a coincidental title of "Practical challenges in quantum key distribution". Are you suggesting that it's a joke and Nature has been pranked :-) $\endgroup$ – Paul Uszak Sep 11 '18 at 17:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.