The quantum cryptography technology allows key distribution in individual segments of a network in a point–point configuration; then, keys in individual segments are used to protect traffic between any nodes of the network, which are not connected directly by a quantum channel. Correspondingly, the problem of negotiation (distribution) of keys between separate segments of the network appears. In this work, the problem of the transmission of an independent key through trusted nodes of the quantum network, between which keys are provided as a result of quantum key distribution, is considered. Quantum keys are used to encrypt the transmitted key. The transmitted key can be encrypted by both the block cipher and the one-time pad. It has been shown that the complexity of search for the key transmitted through the network depends on the imperfection of the external key and quantum keys as well as on the imperfection of the block cipher, i.e., on the average probability of collisions in the block cipher. The complexity in the case of the encryption of the transmitted key by the one-time pad depends only on the imperfection of the transmitted key and encryption keys. Perfect keys keep the perfectness of the transmitted key. In the block cipher, the transmitted key is no longer perfect even with the perfect keys in the measure of difference of the block cipher from the one-time pad. It has also been shown that the larger the number of collisions of the block cipher, i.e., the smaller the number of ciphertexts covered by the block cipher, the smaller the number of search steps required to find the key.
The quantum cryptography technology allows key distribution in individual segments of a network in a point–point configuration; then, keys in individual segments are used to protect traffic between any nodes of the network, which are not connected directly by a quantum channel. Correspondingly, the problem of negotiation (distribution) of keys between separate segments of the network appears. In this work, the problem of the transmission of an independent key through trusted nodes of the quantum network, between which keys are provided as a result of quantum key distribution, is considered. Quantum keys are used to encrypt the transmitted key. The transmitted key can be encrypted by both the block cipher and the one-time pad. It has been shown that the complexity of search for the key transmitted through the network depends on the imperfection of the external key and quantum keys as well as on the imperfection of the block cipher, i.e., on the average probability of collisions in the block cipher. The complexity in the case of the encryption of the transmitted key by the one-time pad depends only on the imperfection of the transmitted key and encryption keys. Perfect keys keep the perfectness of the transmitted key. In the block cipher, the transmitted key is no longer perfect even with the perfect keys in the measure of difference of the block cipher from the one-time pad. It has also been shown that the larger the number of collisions of the block cipher, i.e., the smaller the number of ciphertexts covered by the block cipher, the smaller the number of search steps required to find the key.
The quantum cryptography technology allows key distribution in individual segments of a network in a point–point configuration; then, keys in individual segments are used to protect traffic between any nodes of the network, which are not connected directly by a quantum channel. Correspondingly, the problem of negotiation (distribution) of keys between separate segments of the network appears. In this work, the problem of the transmission of an independent key through trusted nodes of the quantum network, between which keys are provided as a result of quantum key distribution, is considered. Quantum keys are used to encrypt the transmitted key. The transmitted key can be encrypted by both the block cipher and the one-time pad. It has been shown that the complexity of search for the key transmitted through the network depends on the imperfection of the external key and quantum keys as well as on the imperfection of the block cipher, i.e., on the average probability of collisions in the block cipher. The complexity in the case of the encryption of the transmitted key by the one-time pad depends only on the imperfection of the transmitted key and encryption keys. Perfect keys keep the perfectness of the transmitted key. In the block cipher, the transmitted key is no longer perfect even with the perfect keys in the measure of difference of the block cipher from the one-time pad. It has also been shown that the larger the number of collisions of the block cipher, i.e., the smaller the number of ciphertexts covered by the block cipher, the smaller the number of search steps required to find the key.
Abstract The question of the transfer of an external independent key through trusted nodes, between which the keys obtained during the quantum key distribution are located, is investigated. It is shown that when pushing external ɛ K -secret key through a chain of trusted nodes connecting separate segments with ɛ i -secret keys obtained by quantum key distribution, the final matched key between the extreme nodes of the network turns out to be ε = ε K + ∑ i = 1 N s ε i -secret (where i is the number of the network segment, N s is the number of network segments connecting the extreme nodes), provided that the external key is encrypted on the ɛ i –secret keys in the one-time pad mode. It is also shown that for any measurements of an eavesdropper providing him with additional information about the keys, the complexity of partial/complete search for an matched key is determined by the secrecy parameter ɛ .
Abstract Basic configuration in quantum cryptography is a point-to-point quantum key distribution. In networks with quantum key distribution, the keys received on different network segments are different. For information encryption between arbitrary network nodes it is necessary to have agreement of the keys distributed in different segments of network through a public classical communication channel. The secrecy criterion for keys independently distributed in different network segments is formulated in terms of the trace distance. These are the so-called ε –secret keys. The paper deals with the issue of the secrecy of the agreed keys. We propose the method of secret keys agreement in which the agreement of two ε 1 , ε 2 –secret keys distributed in different segments leads to a common ( ε 1 + ε 2 ) –secret key. It is shown that in order to agreed common key will be ε –secret, it is sufficient that the keys on the individual segments are ε /2–secret.
The work is devoted to the study of the problems of using the technology of quantum distribution of secret keys, namely the problems of using and forming new secret keys, based on the generation of quantum keys and the distribution of new keys in networks of complex topologies. In secure communication systems, information is transmitted over public networks, and thus is available to a potential attacker for carrying out various types of attacks. A potential attacker is able to store encrypted data transmitted over network channels, and attempt to decrypt it later, when technical means and attacks on information protection algorithms will allow, in an acceptable time, to carry out decryption. This is the basic principle of "Store now, decrypt later", which is one of the problems of information protection systems. For block ciphers, there is a parameter - the key load, which limits the amount of information that one key can process using the data protection algorithm. NIST recommends a time frame for using secret keys, calculated in years. In the field of information security, there are more significant restrictions on the key load, related to the number of blocks that can be processed by one key, caused by attacks on block ciphers. With these restrictions, there is a need to regularly change the secret key used, and search for options for generating and delivering keys between two network nodes that organize a secure information transmission channel. Changing the secret key to a new key, independent of the previous one, allows you to protect the transfer of confidential information when the key used is compromised. The technology of creating a quantum computer allows you to effectively attack known algorithms for generating secret keys based on the factorization of large numbers or the complexity of calculating a discrete logarithm. The quantum Shor algorithm [5, 6] allows you to solve these problems in polynomial time, so it is necessary to look for alternative solutions to the problem of regular changes in the network channel of the secret key in a pair of nodes. It is important to take into account the need to ensure the quality of Perfect forward secrecy, when the master key is compromised, all subsequent secret keys must not be compromised. To achieve this property of modern encryption algorithms, there is a periodic update of master keys. To prevent the accumulation of information, to successfully carry out attacks by the attacker and to comply with the load requirement on the secret key, it is necessary to solve the problem of regular changes of the secret, and regular generation/delivery of keys to network nodes. Given the urgent needs that are intensively developing today in the secure and high-speed transmission of information, and no less, the rapid growth of technical capabilities, there is a need to develop options in advance that will ensure the security of information during transmission. Regular change, in geographically dispersed data protection systems, of common secret keys, in the case of ensuring the independence of secret keys, allows successfully solving the task.
Quantum key distribution (QKD) protocols generate a same secret key at two nodes which are at different locations. Such key is used to encrypt and decrypt messages for unconditionally secure communications through a classical channel. Originally, QKD protocols are designed to work between two nodes which are directly connected through a point-to-point quantum channel. Currently, commercial QKD systems have limited range and key rate. To increase key rate and extend range, the key generated by QKD at one end node is relayed hop-by-hop to the other end node through multiple trusted nodes. In a QKD network with multiple pairs of communicating nodes, routing deals with the issue of deciding for each pair, which trusted nodes are used to form a path connecting the end nodes. This paper presents an overview of QKD and highlights the differences between QKD routing as compared to classical routing. The challenges of QKD routing are presented. Existing works dealing with the challenges are surveyed. This paper ends with a discussion on remaining research issues.
Secure communication in large-scale quantum key distribution (QKD) networks relies on the key management layer to establish a key between any two nodes in the network. For non-adjacent nodes key forwarding is applied and secure keys are one-time pad encrypted with key material obtained from individual QKD links on a hop-by-hop basis from the initiator to the receiver. Notably, in a distributed setting, this requires that the nodes along the path be fully trusted, since they are given plaintext access to the secret during re-encryption.In this work, we propose a security model for the key forwarding process that provides confidentiality of the key with respect to all nodes that are not on the selected path from initiator to receiver. The model also captures the authenticity of the forwarded key in a sense that the key at the receiver is the same as the one sent by the initiator. Furthermore, in case the authenticity check fails, the receiver can identify the link that violated the authenticity guarantees by accident or on purpose. Finally, we provide a protocol for secure authentic key forwarding based on unconditionally hiding commitment schemes and unforgeable signature schemes that achieves confidentiality of the transmitted keys in an information theoretic sense of security whereas authenticity requires computationally secure primitives.