Abstract:
We experimentally implemented an eavesdropping attack against the Ekert protocol for quantum key distribution based on the Wigner inequality. We demonstrate a serious lack of security of this protocol when the eavesdropper gains total control of the source. In addition we tested a modified Wigner inequality which should guarantee a secure quantum key distribution.

Abstract:
Quantum cryptography (QC) can provide unconditional secure communication between two authorized parties based on the basic principles of quantum mechanics. However, imperfect practical conditions limit its transmission distance and communication speed. Here we implemented the differential phase shift (DPS) quantum key distribution (QKD) with up-conversion assisted hybrid photon detector (HPD) and achieved 1.3 M bits per second secure key rate over a 10-km fiber, which is tolerant against the photon number splitting (PNS) attack, general collective attacks on individual photons, and any other known sequential unambiguous state discrimination (USD) attacks.

Abstract:
We consider the problem of secure key distribution among $n$ trustful agents: the goal is to distribute an identical random bit-string among the $n$ agents over a noisy channel such that eavesdroppers learn little about it. We study the general situation where the only resources required are secure bipartite key distribution and authenticated classical communication. Accordingly, multipartite quantum key distribution can be proven unconditionally secure by reducing the problem to the biparitite case and invoking the proof of security of bipartite quantum key distribution.

Abstract:
The Ekert 91 quantum key distribution (QKD) protocol appears to be secure whatever devices legitimate users adopt for the protocol, as long as the devices give a result that violates Bell's inequality. However, this is not the case if they ignore non-detection events because Eve can make use of the detection-loophole, as Larrson showed. We show that even when legitimate users take into account non-detection events Eve can successfully eavesdrop if the QKD system has been appropriately designed by the manufacturer. A loophole utilized here is that of `free-choice' (or `real randomness'). Local QKD devices with pseudo-random sequence generator installed in them can apparently violate Bell's inequality.

Abstract:
By comparing Cabello's addendum to his quantum key distribution protocol Phys. Rev. A 64(2001)024301], we propose a more convenient modified protocol based on the entanglement swapping which is secure against the eavesdropping strategy addressed by Zhang et al. Phys. Rev. A 63(2001)036301] and other existing types of attack.

Abstract:
In search of a quantum key distribution scheme that could stand up for more drastic eavesdropping attack, I discover a prepare-and-measure scheme using $N$-dimensional quantum particles as information carriers where $N$ is a prime power. Using the Shor-Preskill-type argument, I prove that this scheme is unconditional secure against all attacks allowed by the laws of quantum physics. Incidentally, for $N = 2^n > 2$, each information carrier can be replaced by $n$ entangled qubits. And in this case, I discover an eavesdropping attack on which no unentangled-qubit-based prepare-and-measure quantum key distribution scheme known to date can generate a provably secure key. In contrast, this entangled-qubit-based scheme produces a provably secure key under the same eavesdropping attack whenever $N \geq 16$. This demonstrates the advantage of using entangled particles as information carriers to combat certain eavesdropping strategies.

Abstract:
This paper presents a prepare-and-measure scheme using $N$-dimensional quantum particles as information carriers where $N$ is a prime power. One of the key ingredients used to resist eavesdropping in this scheme is to depolarize all Pauli errors introduced to the quantum information carriers. Using the Shor-Preskill-type argument, we prove that this scheme is unconditionally secure against all attacks allowed by the laws of quantum physics. For $N = 2^n > 2$, each information carrier can be replaced by $n$ entangled qubits. In this case, there is a family of eavesdropping attacks on which no unentangled-qubit-based prepare-and-measure quantum key distribution scheme known to date can generate a provably secure key. In contrast, under the same family of attacks, our entangled-qubit-based scheme remains secure whenever $2^n \geq 4$. This demonstrates the advantage of using entangled particles as information carriers and of using depolarization of Pauli errors to combat eavesdropping attacks more drastic than those that can be handled by unentangled-qubit-based prepare-and-measure schemes.

Abstract:
We prove the security of a quantum key distribution scheme based on transmission of squeezed quantum states of a harmonic oscillator. Our proof employs quantum error-correcting codes that encode a finite-dimensional quantum system in the infinite-dimensional Hilbert space of an oscillator, and protect against errors that shift the canonical variables p and q. If the noise in the quantum channel is weak, squeezing signal states by 2.51 dB (a squeeze factor e^r=1.34) is sufficient in principle to ensure the security of a protocol that is suitably enhanced by classical error correction and privacy amplification. Secure key distribution can be achieved over distances comparable to the attenuation length of the quantum channel.

Abstract:
We demonstrate that a necessary precondition for unconditionally secure quantum key distribution is that sender and receiver can use the available measurement results to prove the presence of entanglement in a quantum state that is effectively distributed between them. One can thus systematically search for entanglement using the class of entanglement witness operators that can be constructed from the observed data. We apply such analysis to two well-known quantum key distribution protocols, namely the 4-state protocol and the 6-state protocol. As a special case, we show that, for some asymmetric error patterns, the presence of entanglement can be proven even for error rates above 25% (4-state protocol) and 33% (6-state protocol).

Abstract:
Secure quantum key distribution can be achieved with an imperfect single-photon source through implementing the decoy-state method. However, security of all those theoretical results of decoy-state method based on the original framework raised by Hwang needs monitoring the source state very carefully, because the elementary proposition that the counting rates of the same state from different sources are equal does not hold in general when the source is unstable. Source intensity monitoring greatly decreases the system efficiency. Here without using Hwang's proposition for stable source, we present a sufficient condition for a secure decoy-state protocol without monitoring the source intensity. The passive 2-intensity protocol proposed by Adachi, Yamamoto, Koashi, and Imoto(AYKI) (Phys. Rev. Lett. 99, 180503 (2007) satisfies the condition. Therefore, the protocol can always work securely without monitoring the source state or switching the source intensity. We also show that our result can greatly improve the key rate of the 3-intensity protocol with a fluctuating coherent-state source.