Long-distance Quantum Communication

The revolutionary scientific results in the 20th century include relativity, quantum mechanics, information theory. From the 1980s, a novel interdiscipline called quantum information has been created by the combination between the two research fields, quantum mechanics and information science. Owing to the potential value for practical application and scientific significance, in recent years quantum information has greatly attracted science and industry community.The fundamental unit of information processing in the classical computer is Bit, i.e., binary string "0" and "1". However, according to one of the basic principle in quantum mechanics, linear superposition principle of quantum states, quantum computer can simultaneously process the arbitrary coherent superposition states of "0" and "1", or qubit. This kind of parallel processing hugely improves the computational speed, and can change some NP problems that cannot be accomplished by classical computers, i.e., the complexity of the problems exponentially increases with the increase of bits, to P problems, i.e., the complexity of the problems polynomially increases with the increase of bits. In 1994 Peter Shor at Bell Laboratory proved that quantum computers could efficiently solve the problem of Large Number Factorization, which was the security footstone of nowaday classical cryptosystems, while none of any classical computers could implement this. Therefore, the discovery of Shor's Algorithm has caused that the classical cryptosystems are not secure at all, and directly threatens the domains of military, government, national security, finance and so on.Fortunately, quantum information also provides another kind of secure communication method, called "quantum cryptography", or quantum key distribution (QKD). The noncloning theorem guarantees the unconditional security. In 1984, Charles Bennett at IBM and Gilles Brassard at Universite de Montreal proposed the first QKD protocol-BB84 protocol, and then Bennett and other co-workers implemented the first experimental demonstration of QKD in 1992. Afterward, especially after the discovery of Shor's Algorithm, due to the significance of quantum information many countries and research communities pay much attention to this research field so that both the theories and experiments of quantum information have greatly developed. Quantum cryptography, as one of the most important sub-branch of quantum communication, has developed towards practical application, and even some corporations have exploited the commercial products of quantum cryptography now.Except for quantum cryptography, there are some other research directions of quantum communication, including single-photon source, entangled source, communication complexity, quantum teleportation, dense coding, entanglement swapping, entanglement purification, quantum repeater, entanglement synchronization, quantum memory, quantum bit commitment, quantum secret sharing, quantum multiparty communication, quantum telecloning and so on. Quantum entanglement is the central resource and plays as a significant role for the research of quantum information. The physical idea of quantum entanglement was originated from Einstein et al.'s doubt on the completeness of quantum mechanics (EPR paradox) in 1935. Although the essences of quantum entanglement are not clear thoroughly, it does not encumber the application of quantum entanglement. On the one hand, quantum entanglement can be used to test the fundamental problem of quantum mechanics such as Bell's inequality, to further understand some counterintuitive phenomena without any analogues in the classical physics such as quantum nonlocality. On the other hand, it can be applied for the quantum information research.Generally, the sources of quantum communication mainly use single-photon source or entangled source while quantum channels utilize free-space channel or optical fiber channel. In this thesis, we will focus on the experimental research of long-distance quantum communication, including experimental entanglement distribution over 13 km, experimental decoy-state quantum key distribution over 102 km using polarization encoding, quantum communication without a shared reference and so on.Furthermore, we will also introduce the experimental demonstration of communication complexity reduction using multiphoton entanglement in a multiparty communication scenario, and the experimental test of All-Versus-Nothing type violation of local realism using two-photon high-dimensional entangled states.Owing to the limitations of current technologies, the maximum distance of fiber quantum communication is limited. In order to implement long-distance quantum communication, there are mainly two kinds of schemes. The first one is the scheme of quantum repeater. The channel between the two communication parties is divided by N segments. In every two neighboring segments, entanglement is independently produced first, and then quantum entanglement can be established between the two segments via entanglement swapping. The quality of entanglement can be improved through quantum purification. Finally entanglement will be created between the two remote parties and preserved using the method of quantum memory. The other method is the free-space scheme. One of the communication parties, say, Alice, distributes the single-photon source or entangled source to Bob through satellites. The major advantage of this scheme is that it can implement quantum communication between arbitrary sites, therefore it is the indispensable way for global quantum communication in the future. Since the attenuation of the free-space channel between ground and satellite is contributed by the aerosphere, its equivalent length is about 5 km and the total attenuation of round-trip is equivalent to 10 km free-space channel in the earth surface aerosphere. The experimental results of free-space entanglement distribution over 13 km validated the feasibility of entanglement based quantum communication between ground and satellite. During the experiment, we also observed the violation of Bell's inequality under the conditions of spacelike interval, which indicated that entanglement could be still maintained over long-distance range.In the case of fiber quantum communication, due to the effect of polarization mode dispersion (PMD) in the single mode fiber and influences of environment in the transmittance channel, the polarization qubits will be decoherent, which is disadvantageous to long-distance fiber quantum communication. In 2005, the Laflamme's group at University of Waterloo, Canada, proposed a feasible scheme to overcome the collective rotation noise of polarization mode in the channel using the so-called "time tag" operations. Experimentally, we implemented this kind of quantum communication scheme without a shared reference to demonstrate robustness of the scheme in the cases of long-distance fiber setting and short-distance fiber setting, respectively.The other central issue for long-distance fiber quantum communication is single-photon source. Since perfect single-photon source does not exist yet, usually weak coherent pulsed light source is used in the experiments. However, this kind of light source exists a significant loophole of security, because Eve can obtain full information of the sender using photon-number-splitting (PNS) attacks so that communication between the two parties is not secure at all. Although there are many theoretical proposals to overcome PNS attacks, decoy-state QKD protocol is one of the most feasible and practical methods. The idea of decoy-state method was first proposed by Hwang in 2003. Subsequently, Xiang-Bin Wang and Lo et al. independently developed this method and finally proposed practical decoy-state QKD schemes. Very recently, our group implemented the experimental one-way decoy-state QKD over 102 km using polarization encoding. It was the first time that the distance of unconditionally secure QKD exceeded 100 km, and paved the way for the practical products of decoy-state quantum cryptosystem in the future.