Quantuminformationisanewly-bornmultidisciplinarysubject,whichcoversseveralclassicalareas,suchascomputerscience,physics,andmathematics. Themajordistinctionof quantum information is known as that tasks, e.g., the information coding, processingandtransmission, canbeachievedusingquantummechanics. Forinstance, quantumcom-puter, which is a combination of quantum mechanics and computer science, and is morepowerfulthanallknownclassicalcomputers,ismakingtroublesforexistingcryptographicsystems. With the popularization of information technology, the demand of informationsecurity is becoming increasingly intensive. Consequently, quantum cryptography wasinvented to meet the requirements.It is known that quantum cryptography,which provides fundamental principles forquantum secure communications, is unconditionally secure in theory. In other words, aneavesdropper,Eve,whoissupposedtobeonlyrestrictedbyquantumlaws,canneverbreakthe system even if she possesses unlimited computation resources. In contrast with math-ematical cryptography, which bases its security on computational complexities, quantumcryptography protects the information in nature, since their security is based on the quan-tum laws, i.e., quantum no-cloning theorem and quantum uncertainty principle. Theo-retically, any eavesdropping in the channel can be detected using quantum cryptography.Therefore, it is definite that quantum cryptography contributes to implementing securecommunications. Up to date, quantum cryptography and quantum communication haveemerged to be a relatively mature system in theory, especially for quantum cryptography,ithasonlybeennomorethan30yearssincethefirstquantumkeydistributionprotocolwasannounced, and now we are able to buy commercial products of quantum cryptographicsystem in the market. Also, various schemes of new protocols and implementations ofquantum cryptography were proposed during these years. Obviously, it can be expectedthat quantum cryptography is becoming one major part of people’s everyday-life in thecoming decades.Up to now, there are still many problems left for further investigations in quantumsecure communications. For example, it is worthy to further study how to devise newquantum cryptography protocols and prove their security, how to implement practical net- works in which quantum cryptography is applied, and how to ensure the security of prac-tical quantum cryptography systems. Motivated by the theoretical interests on quantumsecure communications and accelerating the progress of applying quantum technologiesin real life, also supported by the National Nature Science Foundation of China and theplanning projects of the National University of Defense Technology, this dissertation isdevoted to the investigations of above problems with a new prospect, i.e., informationcounter-working. Generally, information counter-working covers a wide range of topics,among which is the pairwise one, i.e., the information privacy and attacks. Naturally, anew notion, namely quantum information counter-working, is developed by extending theidea of information counter-working into quantum domain. Consequently, the central lineof this dissertation is the topic about quantum cryptography and the attacks on quantumcryptographic systems. Explicitly, it contains sub-topics on the protocols and securityproofs of counterfactual quantum cryptography, continuous-variables quantum cryptog-raphy, quantum communication schemes with high efficiency, and quantum hacking. Thecontributions read as follow:(1) Counterfactual quantum key distribution and quantum coin flippingAs a new paradigm of quantum cryptography, counterfactual quantum key distribu-tion enables two spacially separated parties to share a private key without transmitting anysignal particle. This counter-intuition immediately implies the security, i.e., no key travelsthrough the channel, thus, essentially exhibits security advantages, such as the immunityto the photon number splitting attacks. Existed proof on the security of this new paradigmseems to be unsatisfactory, since the equivalence between the original protocol and thetranslated one should be reconsidered. A new proof, which is more strict than the previ-ous one, is proposed in this dissertation using quantum informatics. At the same time, anew protocol, which outperforms the previous ones on that it produces deterministic keys,is announced.Quantumcoinflippingisanotherhottopicofquantumcryptography. Originally,coinflipping is a cryptographic primitive which allows two separated and untrusted partiesto generate a random bit. Although theoretical proofs have showed that unconditionalsecure coin flipping is impossible, quantum coin flipping does have an advantage overthe conventional one since the later one can be cracked by quantum computers. So far,mostquantumcoinflippingprotocolsareofsingle-shottype, inotherwords, mostofthem produces a single bit as the outcome. However, it should noticed that single-shot quantumcoin flipping protocols might be totally insecure in a real-life channel. To address thisproblem, a new single-shot quantum coin flipping protocol, which is noise-tolerant, isproposed.(2) Continuous-variable quantum cryptography protocolsA new quantum key distribution protocol based on squeezed states is announced, itdistinguishesothersfromthatthestatesaremodulatedwithfouralphabeticletterssymmet-rically, and the modulation is non-gaussian. Two advantages can be found: the securitydoes not degenerate fast with the increase of noise; The efficiency is higher than that ofthe binary-modulated protocols. At the same time, a composed quantum protocol of iden-tity authentication and key distribution is declared using squeezed states. It originates theidea of combining different quantum protocols to improve the practicability of quantumcryptography without any loss of security.(3) Quantum network communication protocols with high efficiencyWiththeincreaseofthenumberofnetworkusers,itisofmuchimportancetoimprovethe communication efficiency of quantum networks. Motivated by this idea, a new quan-tum subnet communication scheme with high efficiency is proposed using quantum datacompression and Grover searching algorithm. Remarkably, this new scheme contributesto bridging the quantum communication and quantum computing algorithms. Quantumnetwork coding is an alternative way to improve the communication efficiency, this dis-sertation has improved the very scheme on quantum network coding with free classicalcommunication. There may be thousands of entangled pairs, upon which quantum tele-portation is available, in a given quantum communication network. Intrinsically, theseentangled pairs are equivalent with hidden channels, the presented scheme is more gen-eral and outperforms the previous ones on the fact that it is still applicable to quantumnetworks with arbitrarily distributed hidden channels. In addition, a network-topology-adaptive quantum secret conference protocol with high efficiency is proposed based ontwo fundamental sub-protocols. Generally, this protocol can be applied to quantum net-works with different topologies.(4) Security analysis of practical quantum cryptography systemsRecently, the subject, namely the security analysis of practical Quantum cryptogra-phy systems, has become one of the hottest subject in quantum cryptography. It generally accomplishes the task of improving the security performance of practical quantum cryp-tography systems by amending their loopholes, which can be employed to launch specificattacks. Our work is presented as follow: First, we explore a new loophole existed in realimplementations of counterfactual quantum cryptography, and devise a new Trojan horseattack, namely the counterfactual attack, based on this new loophole. It is showed thatthe presented attack outperforms all known Trojan horse attacks on that the system can betotally broken even if the imperfection is slight enough. In other words, Eve can extractthe entire key without disturbing the system with our attack. Second, a new time-shift at-tack to real implementations of counterfactual quantum cryptography is announced basedon the reported loophole of avalanche photodiode detectors, it is showed that the reportedimperfection not only threatens the conventional quantum cryptography systems, but alsodegenerates the security of counterfactual ones. At last, a fake-state attack on commercialquantum cryptosystems is proposed, the presented attack is an improvement of two newlyreported attacks on commercial systems. It is showed that the quantum bit error rate canbe reduced using the knowledge of the afterpulsing effect of the avalanche photodiodedetectors. Therefore, the improved attack is more difficult to be detected and easier to beimplemented. |