The EU pursues quantum crypto because of Echelon

David Wagner daw at taverner.cs.berkeley.edu
Tue May 25 16:41:44 EDT 2004


Steve Bellovin  wrote:
>http://www.computerworld.com/securitytopics/security/story/0,10801,93220,00.html?from=homeheads
>
>I'm not sure what more to say, given my opinion of the general utility 
>of quantum crypto....

Based on the way the article is describing it, this sounds like a newly
funded research grant to study quantum crypto.  Grants certainly don't
represent settled policy, and it appears the bit about "defeating Echelon"
may be coming primarily from the researchers getting funded, rather than
from the funders.  So I wouldn't read too much into this news article --
I certainly don't view it as an authoritative statement about the EU's
policy on Echelon and quantum crypto.

I certainly agree with everyone's skepticism about quantum crypto.
Amusingly enough, if quantum crypto was deployed everywhere, it might
indeed reduce the take from satellite surveillance, Echelon, and such
types of SIGINT.  Not because of the quantum crypto, mind you, but
because of the use of fiber optics.  Today's quantum crypto systems
all pretty much require a dedicated fiber optic link between the two
endpoints, and as we know, such links are harder to intercept remotely
than RF traffic.  Consequently, deployment of quantum crypto might reduce
susceptibility to SIGINT by increasing use of dedicated non-RF links
that are hard to tap remotely.  Of course, you'd get basically the same
benefits from sending signals unencrypted over a dedicated, unswitched,
point-to-point fiber optic link, so the quantum crypto is irrelevant.
Also, such dedicated non-RF links are typically very expensive --
they have horrible scaling properties compared to a shared network.
So there are probably far cheaper ways to secure one's infrastructure
against SIGINT, and I'm not going to defend quantum crypto.

---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to majordomo at metzdowd.com



More information about the cryptography mailing list