Based on a photon source, Alice shares with Bob set of photon pulse. But we know that the probability that after a pulse single photon is generated is very small, each pulse will always contain many photons. This might be a threat for any QKD protocol since an eavesdropper (Eve) can intercepts the photon pulse select one photon and sends the remaining to Bob, and waits for Alice and Bob to reveal the measurement bases to perform the measurement on the intercepted photon in the correct bases. This what we call individual attack in QKD. My question is, can entanglement deal with this problem, and if so, how?
$\begingroup$
$\endgroup$
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
This is indeed a problem that can't be ignored when stepping from theory to implementation.
Physical realisations for photon scattering should provide a rate as close as possible to exact single photon emission per logical qubit.
As regard the definition of QKD protocols; there are methods to enhance the security.
The method described in Nielsen and Chuang, Chapter 12.6.2, referred as "Privacy amplification and information reconciliation" is an example.
It is a two-step statistical process that amplifies the ability to rule out Eve, for an arbitrary factor of desired security.
Alice and Bob share two correlated classical strings $X$ and $Y$. Notice that their correlation is due to their source, which can be a string of Bell states (maximally entangled pairs of qubits).
