NEC's quantum telephone

[David Farber <dave () farber net>]

-------- Original Message --------
Subject:        NEC's quantum telephone
Date:   Tue, 22 Nov 2005 22:39:14 +0900
From:   Rod Van Meter <rdv () tera ics keio ac jp>
Reply-To:       rdv () tera ics keio ac jp
To:     dave () farber net



Dave,

[For IP if you wish.  I sent this last week about the end of your email
troubles, so I'm not sure if it got blotted or if it didn't cross
your threshold for "interesting".  Up to you, of course, just making
sure it got the look.]

Chip Elliott and Henry Yeh of BBN were in Tokyo last week, and I helped
introduce them to various quantum computing researchers here, while they
introduced me to some of the quantum key distribution (QKD) folks they
already knew.  You can find some of Chip's papers on the DARPA QKD
network they run at 
http://arxiv.org/find/quant-ph/1/au:+Elliott_C/0/1/0/all/0/1

We went to NEC's Tsukuba facility and, among other things, saw a very
cool demo: NEC's quantum telephone.

More accurately, it's a VoIP phone using QKD-generated keys as a
one-time pad for the conversation.  I'm a little fuzzy on the total
protocol stack, but I don't think they're using full IPSec (unlike the
BBN guys), and they said they haven't implemented the authentication for
the classical channel yet (which is okay for a demo, as long as you're
up front about needing it).

I don't think they're planning on productizing the phone, and they are
coy about productizing the QKD equipment, though it appears to be very
high quality, and they have done a lot of low-level engineering on e.g.
frame synchronization to make it work.  It is, however, a great demo,
and since the final key generation rate is only 13kbps or so over 16km,
it's a good bandwidth match if you're doing one-time pad.

Chip is a long-time Internet and system security guy, so he is very
clear on the value of QKD.  That is, he's skeptical.  He understands
where it fits into the total picture, and also has some ideas about
places where *theory* says QKD is secure, but *practice* might break
down (for example, the detector that detects a one might be more
efficient than the detector that detects a zero; can an eavesdropper
exploit that?).  All in all, BBN's work is very important, IMHO.

I regret to say I didn't get to try it, but I watched Henry and Chip
make a quantum-encrypted phone call.  I'm sorry to report that it
occurred to neither of them to say, "Watson, come here, I need you."

It's not much to look at, and the photo's not great, but I took a
picture of the quantum telephone and put it on my blog at
http://rdvlivefromtokyo.blogspot.com/2005/11/necs-quantum-telephone.html

                --Rod




-------------------------------------
You are subscribed as lists-ip () insecure org
To manage your subscription, go to
 http://v2.listbox.com/member/?listname=ip

Archives at: http://www.interesting-people.org/archives/interesting-people/