Cryptography Reference
In-Depth Information
on one bank of the Danube and stored in a shipping crate. One receiver station,
Alice, was on the far bank of the Danube and located on a rooftop approxi-
mately 500 m away. The second receiver, Bob, was located about 150 m from
the source location on a second rooftop. Although both receiver stations were
located above ground level, the Alice link was periodically blocked by pass-
ing ships, and the Bob link, while not completely blocked, experienced extra
beam fluctuations from passing freight trains on the railroad. Our diode-
laser-pumped down-conversion source requires a fraction of the electrical
power and none of the water cooling of an argon-ion- or titanium-sapphire-
pumped system. For this experiment, all electrical power for the source was
supplied from a gas-powered 2 kW generator. This demonstrates that down-
conversion sources are no longer tied to the laboratory environment and can
be taken virtually anywhere; they can function in real-world applications.
In addition to using the diode-pumped system, we took advantage of a sec-
ond recent advance in entangled photon-pair generation — high efficiency
coupling of the down-conversion light into single-mode optical fiber [73].
Transmission from and collection into nighttime ambient background sources
without resorting to high-loss band-pass filters. The transmitting telescopes
for the experiment were simply single-mode fiber couplers and a 5-cm achro-
matic lens with a 150 mm focal length. The receiver telescopes were identical
except for a polarizer placed in front of the coupler that could be rotated
for polarization measurements. Our singles rate background level was lim-
ited to about 600-700 Hz, which was essentially due only to the dark count-
ing rates of the detectors. The communication schematic for the experiment
is shown in Figure 3.11. Alice's detection signals were sent directly to Bob
via a long coaxial cable that connected the two labs. At the Bob station, a
delay generator was used to account for the extra propagation time of the
Alice signal and synchronize the coincident pulses, which were measured
using standard NIM electronics. While the singles rates and coincidence rates
were measured only at the Bob station, the results were distributed via local
area network (LAN) connections to the Bob rooftop and Wave-LAN to the
source and Alice station. This allowed for remote polarization compensation,
telescope adjustments, and data accumulation using only a single measure-
ment configuration. The source parameters have already been described in
detail in Section 3.2.5. In short, the polarization-entangled singlet Bell state
|
−
1
=
√
2
(
|
HV
−|
VH
)
could be generated with a two-photon visibil-
12
12
12
v
=
.
ity of approximately
95 with singles rates and coincidence rates of ap-
proximately 120,000 Hz and 20,000 Hz, respectively, at a UV pumping power
of 18 mW. The light was coupled through the optical telescope links, each of
which had an attenuation of 12 dB (or about 6% transmission). This was suffi-
cient to yield singles count rates at the receivers of about 4000 s
−
1
(including
background) and a maximum coincidence rate of 15 s
−
1
.
In order to support our claim that the shared photons were entangled,
we measured a set of polarization correlations designed to violate maxi-
mally a CHSH Bell inequality [89,90] for the singlet. We define a polarization
0
