We show that superselection rules do not enhance the information-theoretic security of quantum cryptographic protocols. Our analysis employs two quite different methods. The first method uses the concept of a reference system---in a world subject to a superselection rule, unrestricted operations can be simulated by parties who share access to a reference system with suitable properties. By this method, we prove that if an $n$-party protocol is secure in a world subject to a superselection rule, then the security is maintained even if the superselection rule is relaxed. However, the proof applies only to a limited class of superselection rules, those in which the superselection sectors are labeled by unitary irreducible representations of a compact symmetry group. The second method uses the concept of the format of a message sent between parties---by verifying the format, the recipient of a message can check whether the message could have been sent by a party who performed charge-conserving operations. By this method, we prove that protocols subject to general superselection rules (including those pertaining to non-Abelian anyons in two dimensions) are no more secure than protocols in the unrestricted world. However, the proof applies only to two-party protocols. Our results show in particular that, if no assumptions are made about the computational power of the cheater, then secure quantum bit commitment and strong quantum coin flipping with arbitrarily small bias are impossible in a world subject to superselection rules.
We show that superselection rules do not enhance the information-theoretic security of quantum cryptographic protocols. Our analysis employs two quite different methods. The first method uses the concept of a reference system---in a world subject to a superselection rule, unrestricted operations can be simulated by parties who share access to a reference system with suitable properties. By this method, we prove that if an n-party protocol is secure in a world subject to a superselection rule, then the security is maintained even if the superselection rule is relaxed. However, the proof applies only to a limited class of superselection rules, those in which the superselection sectors are labeled by unitary irreducible representations of a compact symmetry group. The second method uses the concept of the format of a message sent between parties---by verifying the format, the recipient of a message can check whether the message could have been sent by a party who performed charge-conserving operations. By this method, we prove that protocols subject to general superselection rules (including those pertaining to non-Abelian anyons in two dimensions) are no more secure than protocols in the unrestricted world. However, the proof applies only to two-party protocols. Our results show in particular that, if no assumptions are made about the computational power of the cheater, then secure quantum bit commitment and strong quantum coin flipping with arbitrarily small bias are impossible in a world subject to superselection rules.
Bit commitment is an essential primitive in quantum cryptography. It is the basic building block of many cryptographic protocols in which it provides security over distrustful parties. This paper proposes a secured quantum two-bit commitment protocol for any two classical bits. The protocol contains two parties - a committer and a receiver - who share both quantum and classical communication channels. The main phases of this protocol are commitment and revealing phases where the concealing and binding conditions are proved in order to accept the commitment message. The proposed framework utilizes different security layers to hide specific quantum states in a superposition alongside trivial one qubit operations. The secured unitary transformations are applied by the two parties to meet the concealing and binding conditions. To verify the success of the proposed protocol, the measured output of the two parties are compared. The protocol operates under the assumption that the two parties might cheat and the possible existence of an eavesdropper.
Microscopic physical systems requiring a quantum description are increasingly being harnessed to develop useful technologies. The appeal of such an approach lies in the fact that quantum systems possess inherently different properties from those observed on a macroscopic scale. Uniquely quantum phenomena, such as entanglement, offer new tools for performing experimental tasks. For instance, the use of quantum systems for implementing cryptographic protocols and performing computation has caused significant excitement in recent years as it seems likely that the quantum approach offers an advantage over purely classical implementations. The development of quantum technologies leads to the concept of quantum resources. One key quantum resource that lies at the heart of the divergence between quantum and classical descriptions is that of entanglement. However, for entanglement to act as a resource, it must be possible for the parties (we will restrict ourselves to only two) sharing the entanglement to verify its presence. This may seem like a simple task, however, verifying the properties of quantum systems can be nontrivial. To make matters more complicated, there are often constraints under which the parties sharing the quantum systems find themselves which further restrict the ability of the quantum systems to act as a resource. One family of constraints arises when the two parties are dealing with quantum systems which have properties constrained by a conservation law, or some other constraint which limits knowledge of the properties of the local subsystems. These constraints can be expressed using the well studied formalism of superselection rules. When a superselection rule applies to quantum systems it can be ambiguous as to whether entanglement survives between the systems. This thesis considers a situation that arises naturally in a nuclear magnetic resonance experiment where two parties share many copies of identical quantum systems. In this case a superselection rule associated with the finite symmetric group applies due to the indistinguishability of the systems. The concept of a reference frame for the symmetric group is used to alleviate the effect of the superselection rule and demonstrate when entanglement survives despite the constraint of indistinguishability. In addition to this, a situation which goes beyond the superselection formalism is considered, one where the operations performable by the two parties are further constrained. It is shown that despite this stronger constraint, it is still possible in principle to verify the existence of entanglement experimentally as demonstrated by the violation of a Bell inequality. Methods for verifying the presence of entanglement can also prove useful in characterising the properties of entangled quantum systems. Such characterisation may offer insight into the possible advantages offered by quantum systems over their classical counterparts. It becomes possible to more accurately characterise entangled systems by considering a conceptual constraint under which the two parties may find themselves. This is the situation that arises when one of the parties, say Bob, distrusts the other, Alice. In this situation, Alice can only convince Bob of the presence of their entanglement if the entangled system they share possesses a certain property. We term this property steerability, after Schrödinger’s term “steering” [Proc. Camb. Phil. Soc. 31, 553 (1935)] for the Einstein-Podolsky-Rosen effect. This thesis offers a rigorous characterisation of steering in a quantum information context and probes a number of questions relating to this property. First, a number of examples are presented which demonstrate state steerability which is a uniquely quantum property possessed by certain entangled states. Second, the task of experimentally detecting steerability, and thus entanglement, is addressed through the introduction of steerability criteria. Finally two more detailed examples are explored which illustrate various aspects of the measurement dependence of steerability in realistic experiments.
Security trade-offs have previously been established for one-way bit commitment. We study this trade-off in two superselection settings. We show that for an [lsquo]abelian[rsquo] superselection rule (exemplified by particle conservation), the standard trade-off between sealing and binding properties still holds. For the non-abelian case (exemplified by angular momentum conservation), the security trade-off can be more subtle, which we illustrate by showing that if the bit commitment is forced to be ancilla-free, an asymptotically secure quantum bit commitment is possible.
Methods of quantum mechanics promise information-theoretic security for various protocols in cryptography. However, impossibility of some cryptographic applications such as standard bit commitment, oblivious transfer, multiparty secure computations and ideal coin tossing in quantum regime leaves an obvious question on the completeness of quantum cryptography. Instead of using wide range of rules and techniques for a variety of cryptographic applications, we demonstrate here a unified structure for quantum cryptography based on quantum non-local correlations. The unified framework achieves same goals in information-theoretic way as classical cryptography does with computational hardness. To cover the broad range of cryptographic applications, we show that the framework (i) assures secrecy by providing encryption completely unintelligible to eavesdroppers, (ii) guarantees that input from distant parties is concealed unless they are willing to reveal, (iii) assures binding, (iv) allows splitting information between several parties securely and more generally, (v) evades both quantum and classical attacks from internal as well as external eavesdropping.
Coin flipping and bit commitment are two fundamental cryptographic primitives with numerous applications. Quantum information allows for such protocols in the information theoretic setting where no dishonest party can perfectly cheat. The previously best-known quantum coin flipping and bit commitment protocol by Ambainis achieved a cheating probability of at most 3/4 [A. Ambainis, Proceedings of the 30th Annual ACM Symposium on Theory of Computing, Washington, DC, IEEE Computer Society, 2001]. On the other hand, Kitaev showed that no quantum coin flipping or bit commitment protocol can have cheating probability less than 1/2 [A. Kitaev, Presentation at the 6th Workshop on Quantum Information Processing (QIP), 2003]. Closing these gaps has been one of the important open questions in quantum cryptography. In this paper, we resolve both questions. First, we present a quantum strong coin flipping protocol with cheating probability arbitrarily close to 1/2. More precisely, we show how to use any weak coin flipping protocol with cheating probability 1/2+ε in order to achieve a strong coin flipping protocol with cheating probability 1/2+O(ε). The optimal quantum strong coin flipping protocol follows from our construction and the optimal quantum weak coin flipping protocol described by [C. Mochon, arXiv:0711.4114, 2007]. Second, we provide the optimal bound for quantum bit commitment. On the one hand, we show a lower bound of approximately γ≈0.739, improving Kitaev's lower bound. On the other hand, we present an optimal quantum bit commitment protocol which has cheating probability arbitrarily close to γ. More precisely, we show how to use any weak coin flipping protocol with cheating probability 1/2+ε in order to achieve a quantum bit commitment protocol with cheating probability γ+O(ε). To obtain the final protocol, we then use the optimal quantum weak coin flipping protocol described by [C. Mochon, arXiv:0711.4114, 2007]. Unlike the previous protocol for coin flipping, our protocol uses quantum effects beyond the weak coin flip. To stress this fact, we additionally show that any classical bit commitment protocol with access to perfect weak (or strong) coin flipping has cheating probability at least 3/4.