It has been widely claimed and believed that many protocols in quantum key distribution, especially the single-photon BB84 protocol, have been proved unconditionally secure at least in principle, for both asymptotic and finite protocols with realistic bit lengths. In this paper it is pointed out that the only known quantitative justification for such claims is based on incorrect assertions. The precise security requirements are described in terms of the attacker's sequence and bit error probabilities in estimating the key. The extent to which such requirements can be met from a proper trace distance criterion is established. The results show that the quantitative security levels obtainable in concrete protocols with ideal devices do not rule out drastic breach of security unless privacy amplification is more properly applied.
All the currently available unconditional security proofs on quantum key distribution, in particular for the BB84 protocol and its variants including continuous-variable ones, are invalid or incomplete at many points. In this paper we discuss some of the main known problems, particularly those on operational security guarantee and error correction. Most basic are the points that there is no security parameter in such protocols and it is not the case the generated key is perfect with probability
≥1−ε under the trace distance criterion
d≤ε, which is widely claimed in the technical and popular literature. The many serious security consequences of this error about the QKD generated key would be explained, including practical ramification on achievable security levels. It will be shown how the error correction problem alone may already defy rigorous quantitative analysis. Various other problems would be touched upon. It is pointed out that rigorous security guarantee of much more efficient quantum cryptosystems may be obtained by abandoning the disturbance-information tradeoff principle and utilizing instead the known KCQ (keyed communication in quantum noise) principle in conjunction with a new DBM (decoy bits method) principle that will be detailed elsewhere.
All the currently available unconditional security proofs on quantum key distribution, in particular for the BB84 protocol and its variants including continuous-variable ones, are invalid or incomplete at many points. In this paper we discuss some of the main known problems, particularly those on operational security guarantee and error correction. Most basic are the points that there is no security parameter in such protocols and it is not the case the generated key is perfect with probability ≥ 1 - ϵ under the trace distance criterion d ≤ ϵ, which is widely claimed in the technical and popular literature. The many serious security consequences of this error about the QKD generated key would be explained, including practical ramification on achievable security levels. It will be shown how the error correction problem alone may already defy rigorous quantitative analysis. Various other problems would be touched upon. It is pointed out that rigorous security guarantee of much more efficient quantum cryptosystems may be obtained by abandoning the disturbance-information tradeoff principle and utilizing instead the known KCQ (keyed communication in quantum noise) principle in conjunction with a new DBM (decoy bits method) principle that will be detailed elsewhere.
We analyze the fundamental security significance of the quantitative criteria on the final generated key
K in quantum key generation including the quantum criterion
d, the attacker's mutual information on
K, and the statistical distance between her distribution on
K and the uniform distribution. For operational significance a criterion has to produce a guarantee on the attacker's probability of correctly estimating some portions of
K from her measurement, in particular her maximum probability of identifying the whole
K. We distinguish between the raw security of
K when the attacker just gets at
K before it is used in a cryptographic context and its composition security when the attacker may gain further information during its actual use to help get at
K. We compare both of these securities of
K to those obtainable from conventional key expansion with a symmetric key cipher. It is pointed out that a common belief in the superior security of a quantum generated
K is based on an incorrect interpretation of
d which cannot be true, and the security significance of
d is uncertain. Generally, the quantum key distribution key
K has no composition security guarantee and its raw security guarantee from concrete protocols is worse than that of conventional ciphers. Furthermore, for both raw and composition security there is an exponential catch-up problem that would make it difficult to quantitatively improve the security of
K in a realistic protocol. Some possible ways to deal with the situation are suggested.
It is claimed in the many papers that a trace distance (
d) guarantees the universal composition security in quantum key distribution (QKD). In this introduction paper, at first, it is explicitly explained what is the main misconception in the claim of the unconditional security for QKD theory. In general terms, the cause of the misunderstanding on the security claim is the Lemma in the paper of Renner. It suggests that the generation of the perfect random key is assured by the probability $ (1-d)$, and its failure probability is
d. Thus, it concludes that the generated key provides the perfect random key sequence when the protocol suceeds. So the QKD provides perfect secrecy to the one time pad. This is the reason for the composition claim. However, the quantity of the trace distance (or variational distance) is not the probability for such an event. If $d $ is not small enough, always the generated key sequence is not uniform. Now one needs the reconstruction of the evaluation of the trace distance if one wants to use it. One should first go back to the indistinguishability theory in the computational complexity based, and to clarify the meaning of the value of the variational distance. In addition, the same analysis for the information theoretic case is necessary. The recent serial papers by H.P.Yuen have given the answer on such questions. In this paper, we show more concise description of Yuen's theory, and clarify that the recent upper bound theories for the trace distance by Tomamichel et al and Hayashi et al are constructed based on the reasoning of Renner and it is unsuitable as the analysis for information theoretic security. Finally, we introduce a macroscopic quantum communication with different principle to replace Q-bit QKD.
The security significance of the trace distance security criterion
d is analyzed in terms of operational probabilities of an attacker's success in identifying different subsets of the generated key, both during the key generation process and when the key is used in one-time pad data encryption under known-plaintext attacks. The difference between Eve's sequence error rate and bit error rate is brought out. It is shown with counter-examples that the strong security claim maintained in the literature is incorrect. Other than the whole key error rates that can be quantified at the levels d^{1/3} and d^{1/4} which are much worse than
d itself, the attacker's success probabilities in estimating various subsets of the key and in known-plaintext attacks are yet to be quantified from
d if possible. It is demonstrated in realistic numerical examples of concrete protocols that drastic breach of security cannot yet be ruled out.