Can Eve repeat?

[Ian Grigg]

"R. Hirschfeld" wrote:
> 
> > Date: Fri, 19 Sep 2003 11:57:22 -0400
> > From: Ian Grigg <[EMAIL PROTECTED]>
> 
> > If I understand this correctly, this is both
> > an eavesdropping scenario and an MITM scenario.
> >
> > In the above, Eve is acting as Mallory, as she
> > is by definition intercepting the bits and re-
> > sending them on?
> 
> As Dave Howe pointed out, Eve is acting as a repeater and tries not to
> alter the bits.  This seems a sensible model of eavesdropping for QKD.
> The threat is that Alice and Bob might incorporate bits that were seen
> by Eve into their key.  If Bob never receives a bit, it won't be used.

I'm curious on one point of terminology - my understanding
of the term "EVE" was that she could not be active, and
could not repeat.  Is this defined anywhere?

I had a look at the big red book, but Schneier doesn't
quite nail that issue.  He seems to err on the side of
Eve not being able to repeat, as the dividing line is
set as active v. passive.

As I say, just a point of terminology.

iang

-
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]

Re: quantum hype

[R. Hirschfeld]

> Date: Fri, 19 Sep 2003 11:57:22 -0400
> From: Ian Grigg <[EMAIL PROTECTED]>

> If I understand this correctly, this is both
> an eavesdropping scenario and an MITM scenario.
> 
> In the above, Eve is acting as Mallory, as she
> is by definition intercepting the bits and re-
> sending them on?

As Dave Howe pointed out, Eve is acting as a repeater and tries not to
alter the bits.  This seems a sensible model of eavesdropping for QKD.
The threat is that Alice and Bob might incorporate bits that were seen
by Eve into their key.  If Bob never receives a bit, it won't be used.

> That is, the "Quantum Property" is that Eve can
> be detected because she destroys photos in the
> act of listening, and Mallory, who can resend
> the photons, has only a 50% chance of reading
> each bit correctly in advance, so he can be
> detected after the fact as well, as 25% of his
> bits are wrong.

The terminology "destroy" is used a bit loosely.  I think the
important thing for QKD is that if a photon is measured with the wrong
basis, the information it is carrying about the key is lost.

Ray

-
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]