We characterize a noisy variant of a Quantum public encryption protocol recently introduced by Malavolta and Walter which demonstrated that the notion of everlasting security can be rigorously formulated for Quantum key distribution after two rounds of interaction between Alice and Bob. To address one possible direction of research that is related to injecting noise in the cryptographic protocol related to Quantum key distribution we formulate arguments for further examining the notion of everlasting security through malleability assumptions on transformations of the ciphertext. Assumptions surrounding malleability were introduced by Maurer and Tackmann for the purposes of comparing how authenticate then encrypt, and encrypt then authenticate, protocols behave through a variety of expressions for the forwarding error, deleting error, and reconstruction probabilities. Such probabilities are put to further use for obtaining connections between the indistinguishability and security threshold for a cryptographic protocol of interests. To further build upon such associations we demonstrate, through an adaptation of the Gentle Measurement Lemma from Quantum information theory, how upper bounds on the trace distance can be used to generalize the negligibility function obtained by Malavolta and Walter in the noiseless setting. Besides the fact that the negligibility function in the noisy setting is related to a higher security threshold it continues to remain of interest to determine whether computations provided in this work for upper bounding the trace distance can be related to other settings that are centered more on game-theoretic approaches.
We characterize a noisy variant of a Quantum public encryption protocol recently introduced by Malavolta and Walter which demonstrated that the notion of everlasting security can be rigorously formulated for Quantum key distribution after two rounds of interaction between Alice and Bob. To address one possible direction of research that is related to injecting noise in the cryptographic protocol related to Quantum key distribution we formulate arguments for further examining the notion of everlasting security through malleability assumptions on transformations of the ciphertext. Assumptions surrounding malleability were introduced by Maurer and Tackmann for the purposes of comparing how authenticate then encrypt, and encrypt then authenticate, protocols behave through a variety of expressions for the forwarding error, deleting error, and reconstruction probabilities. Such probabilities are put to further use for obtaining connections between the indistinguishability and security threshold for a cryptographic protocol of interests. To further build upon such associations we demonstrate, through an adaptation of the Gentle Measurement Lemma from Quantum information theory, how upper bounds on the trace distance can be used to generalize the negligibility function obtained by Malavolta and Walter in the noiseless setting. Besides the fact that the negligibility function in the noisy setting is related to a higher security threshold it continues to remain of interest to determine whether computations provided in this work for upper bounding the trace distance can be related to other settings that are centered more on game-theoretic approaches.
Non-malleability is an important security property for public-key encryption (PKE). Its significance is due to the fundamental unachievability of integrity and authenticity guarantees in this setting, rendering it the strongest integrity-like property achievable using only PKE, without digital signatures. In this work, we generalize this notion to the setting of quantum public-key encryption. Overcoming the notorious "recording barrier" known from generalizing other integrity-like security notions to quantum encryption, we generalize one of the equivalent classical definitions, comparison-based non-malleability, and show how it can be fulfilled. In addition, we explore one-time non-malleability notions for symmetric-key encryption from the literature by defining plaintext and ciphertext variants and by characterizing their relation.
The trace distance criterion for security of Quantum Key Distribution (QKD) has been widely perceived that its upperbound is the maximum failure probability in distributing an ideal quantum state that the quantum system shared by legitimate users is decoupled from the quantum system the eavesdropper possesses, therefore the eavesdropper would not obtain any hints on the key shared by the legitimate users no matter how much her measurement is optimal. However, there an arbitrariness in the definition of the trace distance with the decoupled quantum system the eavesdropper possesses as the previous work pointed out, and such an arbitrariness would be confusions in guaranteeing the security of QKD. In this work, from the framework of Bit-Error-Rate (BER) Guarantee discussed also in the previous work, such an arbitrariness is removed from the definition of the trace distance criterion by discussing the similarities and differences between the security of QKD and classical information-theoretic cryptography by Leftover Hash Lemma (LHL). However, in the previous work BER Guarantee was derived under the assumption that the eavesdropper launches the weakest class of attack, so-called “Individual Attacks.” Therefore, it is still unknown whether direct upper-bounding of the term would give tighter upper-bound in general, such as “Collective Attacks” or “Coherent Attacks”, compared to the upper-bound by LHL, as well as the secure-key generation rate. This study also revisits the security proof of QKD given by P. Shor and J. Preskill in 2000.
We examine public broadcast, forward conceptual, and backward conceptual, Quantum channels in the context of communication protocols that are independent of secret keys. Given research directions of interest previously identified in arXiv: 1804.01797, besides converse upper bounds on the bit transmission rate obtained by the author in recent work (arXiv: 2507.03035), additional possibilities remain, including: (1) determining whether aspects of QKD dependent protocols can be incorporated into steps of QKD independent protocols; (2) whether there would be any amplification to the Quantum-classical performance gap that Alice and Bob can exploit towards prospective Quantum advantage; (3) formulating the conditions under which secrecy and authentication can be simultaneously achieved. To characterize the conditions for which secrecy can be achieved with high probability, we argue that there not only exists suitable protocols which enable Alice and Bob to map into the authenticated space of bit codewords with high probability, but also that forward conceptual channels, through cascading, can significantly increase Eve's probability of false acceptance. Albeit the fact that secrecy, along with conceputalizations of the public broadcast channel, were initially discussed by Maurer for QKD dependent protocols, determining whether aspects of such protocols can be adapted for unconditional security without the use of a secret key is of great interest to explore. We demonstrate that Eve's error probability, through the cascading procedure, can be analyzed with the Holevo information under an optimal decoder. Furthermore, through post-processing of the outputs of a Completely Positive Trace Preserving (CPTP) map, we also demonstrate how to decrease Holevo sum quantities with data-processing and entropy-continuity bounds.
This thesis presents a formal model for quantum protocols, that is, cryptographic protocols based on quantum mechanics. It also develops a definition of security (which includes notions such as privacy and correctness) for quantum protocols.
First, existing classical (non-quantum) definitions of security are studied. There are essentially two approaches: one based on Turing Machines and simulation, the other using information theory. Both approaches are explored in detail and compared, then a particular variation of the first approach is chosen. Security is defined by introducing an interface (also called a simulator) which is capable of translating an attack on protocol 1 into an attack on protocol 2. The profile of a protocol is a stochastic variable, defined as the output of the participants and the view of the adversary who performs the attack. If an interface exists such that the profile of protocol 1 is indistinguishable from the profile of protocol 2 with the help of this interface, then protocol 1 is at least as secure as protocol 2. An absolute notion of security is obtained by comparing the protocol implementation with its 'ideal' equivalent.
In this approach, and in classical cryptography in general, indistinguishability of probability distributions is a crucial tool. The meaning of this notion is explored in the context of density matrices, the mathematical formalism which is used to describe the state of a quantum system. Various measures of distinguishability of quantum states relevant from a cryptographic point of view are investigated: probability of error, trace norm (related to the Kolmogorov distance), overlap (related to the Bhattacharyya coefficient) and Shannon distinguishability (defined through the mutual information). It is shown that if we require that the distinguishability between two families of quantum states vanish exponentially fast, these four distinguishability measures are equivalent.
Along the way upper and lower bounds for the Shannon distinguishability are obtained. This is a very useful result, because explicit computation of the Shannon distinguishability between two quantum states is often difficult (it requires solving a transcendental equation), whereas the other distinguishability measures, in particular the trace norm, are usually much easier to compute.
In the last chapter the two previous topics are combined. A formal model for quantum protocols is formulated. It uses Quantum Turing Machines to represent the participants, who communicate through a quantum channel. The different nature of quantum information as opposed to classical information is explored in detail, including the implications this has for the definition. As in the classical case, we introduce an interface to compare the security of two protocols for the same task. Subsequently some consequences of the proposed definition are investigated. In particular, strong evidence is given that one classical technique often used in proofs of security for classical protocols, called rewinding, is impossible in the quantum setting.
We study two novel paradigms in quantum error correction and quantum cryptography — approximate quantum error correction and noisy-storage cryptography — which explore alternate approaches for dealing with quantum noise. Approximate quantum error correction seeks to relax the constraint of perfect error correction and construct codes that might be better adapted to correct for specific noise models. Noisy-storage cryptography relies on the power of quantum noise to execute two-party cryptographic tasks securely. Motivated by examples of approximately correcting codes, which make use of fewer physical resources than perfect codes and still obtain comparable levels of fidelity, we study the problem of finding and characterizing such codes in general. We construct for the first time a universal, near-optimal recovery map for approximate quantum error correction (AQEC), with optimality defined in terms of worst-case fidelity. Using the analytical form of this recovery, we also obtain easily verifiable conditions for AQEC. This in turn leads to a simple algorithm for identifying good approximate codes, without having to perform a difficult optimization over all recovery maps for every possible encoding. Noisy-storage cryptography envisions a setting where two-party cryptographic protocols can be securely implemented based solely on the assumption that the quantum storage device possessed by either party is noisy and bounded. Here, we construct two-party protocols (using higher-dimensional states) that are secure even when a dishonest player can store all but a small fraction of the information transmitted during the protocol, in his noiseless quantum memory. We also show that when his memory is noisy, security can be extended to a larger class of noisy quantum memories. Our result demonstrates that the physical limits of the quantum noisy-storage model are indeed achievable, albeit asymptotically. We also describe our investigations on obtaining strong entropic uncertainty relations using symmetric complementary bases. Uncertainty relations are an important and useful resource in analyzing the security of quantum cryptographic protocols, in addition to being of interest from a foundational standpoint. We present a novel construction of sets of symmetric, complementary bases in dimension d = 2n, which are cyclically permuted under the action of a unitary transformation. We also obtain new lower bounds for uncertainty relations in terms of the min-entropy, which are tight for specific instances of our construction.
We obtain estimates for Eve's forgery probability, namely the probability that she is able to forge a message which Alice or Bob mistakenly accept over a noisy Quantum channel for generating a shared Quantum secret key. This probability is related to Eve's success probability obtained in a previous work due to Renner and Wolf, which was obtained from assumptions on the min-entropy for characterizing asymmetric security. To demonstrate that protocols over noisy Quantum channels are dependent upon a single, unified security threshold in comparison to multiple security parameters in the Renner-Wolf interactive authentication protocol framework we upper bound Eve's forgery probability with a Holevo-type quantity that can be made negligibly small. By leveraging estimates for Eve's false acceptance probability that have previously been obtained by the author, we obtain the desired security threshold by bounding the false acceptance probability with a suitably chosen two-universal function which serves as a counterpart to two-universal hashing functions that have previously been examined for cryptographic protocols in Quantum key distribution. As a result the protocol is not only ε-secure, for some ε>0, but also composable against forgery and key leakage.