Efficient Quantum Secure Direct Communication with Authentication

Two protocols of quantum direct communication with authentication [Phys. Rev. A 73 （2006） 042305] were recently indicated to be insecure against the authenticator Trent attacks [Phys. Rev. A 75 （2007） 026301]. We present two efficient protocols by using four Panli operations, which are secure against inner Trent attacks as well as outer Eve attacks. Finally, we generalize them to multiparty quantum direction communication.     

High Efficiency of Two Efficient QSDC with Authentication Is at the Cost of Their Security

Two efficient protocols of quantum secure direct communication with authentication [Chin. Phys. Lett. 25 （2008） 2354] were recently proposed by Liu et al. to improve the efficiency of two protocols presented in [Phys. Rev. A 75 （2007） 026301] by four Pauli operations. We show that the high efficiency of the two protocols is at the expense of their security. The authenticator Trent can reach half the secret by a particular attack strategy in the first protocol. In the second protocol, not only Trent but also an eavesdropper outside can elicit half-information about the secret from the public declaration.     

Reexamining the security of the improved quantum secret sharing scheme

We find that, in the improvement [S.J. Qin et al., Phys. Lett. A 357 (2006) 101] of the multiparty quantum secret sharing [Z.J. Zhang et al., Phys. Rev. A 71 (2005) 044301], Charlie can solely obtain Alice’s secret messages without Bob’s helps. In other words, the improved secret sharing scheme is still insecure. In the end, we further modify Qin et al. improved three-party quantum secret sharing scheme and make it really secure.     

Threshold Quantum Secret Sharing of Secure Direct Communication

We propose a （L, n）-threshold quantum secret sharing protocol of secure direct communication following some ideas of Zhang＇s protocol [Phys. Lett. A 342 （2005） 60] and Tokunaga et al.＇s protocol [Phys. Rev. A 71 （2005） 012314]. The sender distributes the classical secret shares to his or her n agents and each agent owns a secret share in advance. The sender＇s secure direct communication message can be extracted by an agent subset by collaboration in such a way that at least t or more agents can obtain the secret message with the mutual assistances but any t - 1 or fewer agents cannot. In contrast to the previous multiparty quantum secret sharing protocols in which the sender＇s secret message can be recovered only if all the agents collaborate, our protocol is more practical and more flexible.     

Multiparty quantum secret sharing based on the improved Boström-Felbinger protocol

Based on the famous quantum secure direct communication protocol (i.e., the Boström-Felbinger protocol) [Phys. Rev. Lett. 89 (2002) 187902] and its improvements, we propose a scheme of multiparty quantum secret sharing of classical messages (QSSCM), in which no subset of all the classical message receivers is sufficient to extract the sender’s secret classical messages but all the parties cooperate together. Then we take advantage of this multiparty QSSCM scheme to establish a scheme of multiparty secret sharing of quantum information (SSQI), in which the unknown quantum state in the sender’s qubit can be reconstructed in one receiver’s qubit if and only if all the quantum information receivers collaborate together.     

A Modified Protocol of Quantum Key Distribution Based on Entanglement Swapping

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.     

Improvement of Security of Three-Party Quantum Secure Direct Communication Based on GHZ States

A protocol for three-party quantum secure direct communication based on Greenberger-Horne-Zeilinger （GHZ） states was recently proposed by 3in et al. [Phys. Lett. A 354 （2006） 67] By analysing the protocol we find some security loopholes, e.g. one bit of secret messages of a party （Alice in the original paper） can always be leaked straight to the public without any eavesdropping. These problems suggested previously are discussed and possible solutions are presented to improve the security of the original protocol.     

Efficient Multiparty-to-Multiparty Quantum Secret Sharing via Continuous Variable Operations

The one-to-multiparty quantum secret sharing scheme [Phys. Rev. A 71 （2005） 044301] proposed recently is extended to a multiparty-to-multiparty case. Furthermore, the continuous variable operations are employed in the extended scheme to replace the specific discrete unitary operations used in the original scheme. The complete randomicity of the continuous variable characterizing the unitary operations can ensure the security of secret sharing. Moreover, the present scheme is compared with the recent similar scheme [Phys. Rev. A 72 （2005） 012304]. It is found that the efficiency of the present scheme is n times of that of the previous one.     

Comment on: “Three-party quantum secure direct communication based on GHZ states” [Phys. Lett. A 354 (2006) 67]

The GHZ-state-based quantum secure direct communication (QSDC) protocol [X.-R. Jin, et al., Phys. Lett. A 354 (2006) 67] and its improved version [Z. Man, Y. Xia, Chin. Phys. Lett. 24 (2007) 15] are analyzed from the aspect of security. It shows that much information of the transmitted secret message will be leaked out in both protocols.     

On the optimality of individual entangling-probe attacks against BB84 quantum key distribution

Some MIT researchers [Phys. Rev. A 75, 042327 (2007)] have recently claimed that their implementation of the Slutsky-Brandt attack [Phys. Rev. A 57, 2383 (1998); Phys. Rev. A 71, 042312 (2005)] to the BB84 quantum-key-distribution (QKD) protocol puts the security of this protocol “to the test” by simulating “the most powerful individual-photon attack” [Phys. Rev. A 73, 012315 (2006)]. A related unfortunate news feature by a scientific journal [G. Brumfiel, Quantum cryptography is hacked, News @ Nature (april 2007); Nature 447, 372 (2007)] has spurred some concern in the QKD community and among the general public by misinterpreting the implications of this work. The present article proves the existence of a stronger individual attack on QKD protocols with encrypted error correction, for which tight bounds are shown, and clarifies why the claims of the news feature incorrectly suggest a contradiction with the established “old-style” theory of BB84 individual attacks. The full implementation of a quantum cryptographic protocol includes a reconciliation and a privacy-amplification stage, whose choice alters in general both the maximum extractable secret and the optimal eavesdropping attack. The authors of [Phys. Rev. A 75, 042327 (2007)] are concerned only with the error-free part of the so-called sifted string, and do not consider faulty bits, which, in the version of their protocol, are discarded. When using the provably superior reconciliation approach of encrypted error correction (instead of error discard), the Slutsky-Brandt attack is no more optimal and does not “threaten” the security bound derived by Lütkenhaus [Phys. Rev. A 59, 3301 (1999)]. It is shown that the method of Slutsky and collaborators [Phys. Rev. A 57, 2383 (1998)] can be adapted to reconciliation with error correction, and that the optimal entangling probe can be explicitly found. Moreover, this attack fills Lütkenhaus bound, proving that it is tight (a fact which was not previously known).     

Quantum Authenticated Direct Communication Using Bell States

Two protocols of quantum direct communication with authentication [Phys. Rev. A 73:042305, 2006] were recently proposed by Lee, Lim and Yang, based on the correlation of Greenberger-Horne-Zeilinger (GHZ) states. However, Zhang et al. showed that in the two protocols the authenticator Trent can eavesdrop the secret message by subtle strategies [Phys. Rev. A 75:026301, 2007]. In this paper, we propose two authenticated quantum direct communication (AQDC) protocols using Bell states. Users can identify each other by checking the correlation of Bell states. Alice can directly send a secret message to Bob without any previously shared secret using the remaining Bell states after authentication. The two proposed AQDC protocols are implemented under the condition that there is a quantum link between Alice and Bob and that there is no quantum link between Alice and Bob respectively, similar to the ones proposed by Lee, Lim and Yang [Phys. Rev. A 73:042305, 2006]. The proposed AQDC protocols not only fix the leaks in the AQDC protocols proposed by Lee, Lim and Yang, but also economize the quantum resource.     

Method for generating a new class of multipartite entangled state in cavity quantum electrodynamics

Recently, Yeo and Chua introduced a genuine four-qubit entangled state |χ〉 which can implement perfect teleportation of an arbitrary two-qubit state [Y. Yeo, W.K. Chua, Phys. Rev. Lett. 96 (2006) 060502]. It has been shown that the state |χ〉 is inequivalent to the well-known Greenberger-Horne-Zeilinger state, W state, and linear cluster state, in terms of stochastic local operations and classical communication [C.F. Wu, Y. Yeo, L.C. Kwek, C.H. Oh, Phys. Rev. A 75 (2007) 032332]. This “new” class of state has many interesting entanglement properties and possible applications in quantum-information processing and fundamental tests of quantum mechanics. Here, we propose a simple scheme to generate the state |χ〉 in cavity quantum electrodynamics. Our idea may be helpful for in-depth study on such a class of state and its practical applications.     

Probabilistic generalized controlled order rearrangement encryption and secret splitting using W-class state

Based on the controlled order rearrange encryption (CORE) for quantum key distribution using EPR pairs [Fu.G. Deng, G.L. Long, Phys. Rev. A 68 (2003) 042315], we propose a generalized controlled order rearrangement encryption (GCORE) protocol using non-maximally entangled W-class states with probability, but it also has full efficiency and we compare the similarity and difference with original protocol. Besides, we use this W-class state to split quantum information, thus the scheme is robust against decoherence.     

A genuine four-partite entangled state

In a recent paper, a genuine four-partite entangled state is proposed [Y. Yeo, W.K. Chua, Phys. Rev. Lett. 96 (2006) 060502], which has been found to have many interesting entanglement properties. We show this state is locally equivalent to some graph states.     

Class of Unlockable Bound Entangled States and Their Applications

A class of unlockable bound entangled states and their applications are presented. They can be considered as quasi generalized Smolin states [Phys. Rev. A 63 （2001） 032306], which are the states of N ＋ N qubits. No pure entanglement can be distilled from this class of states by local quantum operations and classical communications. However, if certain parties group together, they become distillable. Although they are bound entangled states, they could be used to achieve some non-trivial tasks, such as quantum secret sharing shown in the study.     

Re-examining generalized teleportation protocol

We present an explicit generalized protocol for probabilistic teleportation of an arbitrary N-qubit GHZ entangled state via only one non-maximally two-qubit entangled state. Without entanglement concentration, using standard Bell-state measurement and classical communication one cannot teleport the state with unit fidelity and unit probability. We show that by properly choosing the measurement basis it is possible to achieve unity fidelity transfer of the state. Compared with Gordon et al’s protocol [G. Gordon, G. Rigolin, Phys. Rev. A 73 (2006) 042309], this protocol has the advantage of transmitting much less qubits and classical information for teleporting an arbitrary N-qubit GHZ state.     

Cryptanalysis and Improvement of a Quantum Secret Sharing Protocol between Multiparty and Multiparty with Single Photons and Unitary Transformations

In a recent paper [Chin. Phys. Lett 25（2008）1187], a quantum secret sharing scheme between multiparty and multiparty was presented. We show that the protocol is not secure because the last member in Alice＇s group can illegally obtain most secret messages without introducing any error. Finally, a possible way to avoid the security flaw is suggested.     

An improved quantum key distribution protocol based on second-order coherence

We improve the quantum key distribution protocol proposed by Pereira et al. [S.F. Pereira, Z.Y. Ou, H.J. Kimble, Phys. Rev. A 62 (2000) 042311], by employing the second-order coherence of optical fields, which can be easy experimentally measured with a Hanbury-Brown and Twiss intensity interferometer. It is shown that eavesdropping can be directly detected without sacrificing extra secret bits as test key. The efficiency of the improved system is enhanced greatly, since no secret bit needs to be discarded.     

Linear optical implementation of a quantum network for quantum estimation

We present an interferometer for simulating the quantum network for quantum estimation proposed by A.K. Ekert et al. [A.K. Ekert, C.M. Alves, D.K.L. Oi, M. Horodecki, P. Horodecki, L.C. Kwek, Phys. Rev. Lett. 88 (2002) 217901]. We experimentally perform overlap measurements of two single-qubit states with linear optical elements. The scheme is generalized to perform estimation of Trρ³.     

Exposed-key weakness of αη

The αη protocol given by Barbosa et al. [G.A. Barbosa, E. Corndorf, P. Kumar, H.P. Yuen, Phys. Rev. Lett. 90 (2003) 227901, quant-ph/0212018] claims to be a secure way of encrypting messages using mesoscopic coherent states. We show that transmission under αη exposes information about the secret key to an eavesdropper, and we estimate the rate at which an eavesdropper can learn about the key. We also consider the consequences of using further randomization to protect the key and how our analysis applies to this case. We conclude that αη is not informationally secure.