Cryptography Reference
In-Depth Information
10.1 Introduction
Of all the capabilities afforded by quantum information science [1], quantum
key distribution (QKD; for a review, see Reference [2]) currently shows the
most promise for practical implementation. Accordingly, there has been a con-
certed effort to develop QKD schemes that mitigate the technical challenges
associated with existing approaches. Among the successes in this effort are the
development of noise-immune (alignment-free) schemes for polarization [3]
and time-bin [4-7] qubits. A further advance is the development of a sym-
metric scheme for time-bin qubits in which neither Alice nor Bob is required
to make active changes to their setups [8]. Here we use the term symmetric to
describe QKD schemes in which a central source distributes some number of
photons to both Alice and Bob, so that they share entanglement. This is in con-
trast to round-trip and one-way configurations, in which the photons move
according to Bob
Bob, respectively. Here we show
that symmetry and noise-immunity can be combined in a single implementa-
tion, for both polarization and time-bin qubits. Beginning with polarization-
coded QKD, we first present a round-trip scheme in which noise-immunity is
achieved by sampling the channel birefringence twice (once on the way from
Bob to Alice and once on the way back). Second, we show how Klyshko's “ad-
vanced wave interpretation” (AWI) [9] can be used to transform this round-
trip scheme into a one-way scheme imbued with passive detection. Third, we
apply the AWI again to obtain a symmetric noise-immune scheme in which
both Alice and Bob have passive setups. We then repeat these three steps for
time-bin-coded QKD. For each scheme, we present a feasible implementation
that relies only on current technology.
→
Alice
→
Bob, and Alice
→
10.2 Noise-Immune Polarization-Coded
Schemes
10.2.1 Round-Trip Noise-Immune
Polarization-Coded QKD
The left column of Figure 10.1 shows the space-time diagrams of three noise-
immune polarization-coded QKD schemes. For polarization qubits, noise-
immune means that the scheme is immune to channel birefringence. The first
scheme [Figure 10.1(A)] requires a round trip and is active (both Alice and
Bob are required to make changes to their respective setups). The scheme
runs as follows. Bob randomly chooses between polarization states
|
V
and
|
(here, and for the rest of this chapter, we suppress normalization
factors) and sends a single photon in that state to Alice. Alice uses a Faraday
mirror to reflect that single photon back, and she also sends along an auxil-
iary unpolarized photon. Alice encodes a single bit by controlling the time
ordering of the two photons she sends to Bob. Bob then measures each pho-
ton in the basis associated with the state of the initial photon he sent. Without
knowing which state Bob sent to Alice, Eve cannot deterministically learn
H
+|
V
