Mina Doosti

Physical Unclonable Functions (PUFs) are physical devices that have unique behaviour which is hard to clone. These hardware structures are considered as an effective and feasible security primitive. The application of a wide variety of PUF structures for different security purposes such as identification and key generation has been widely studied in the context of Classical PUFs. In addition, the quantum-readout PUF (QR-PUF) has been studied as a proposition for a quantum version of classical PUFs. In this paper, we do a comprehensive study on Quantum Physical Unclonable Functions with quantum cryptographic tools. We use a quantum game-based security framework for our analysis and we define a new class of quantum attacks, called General Quantum Emulation Attack (GQEA), applicable on current quantum-readout and hybrid quantum-classical PUFs. This class of attacks are based on using a database of inputs and outputs to emulate the action of an unknown quantum transformation on a new input. We define a concrete attack based on an existing emulation algorithm and use it to show the vulnerability of the current schemes under this attack. Furthermore, we formally define a QPUF for the first time and discuss the security of Unitary QPUFs (UQPUFs) by formally defining the unforgeability property of UQPUFs. We prove any UQPUF provides selective unforgeability property while they cannot provide unconditional and existential unforgeabilities.

Known protocols for the secure delegation of quantum computations from a client to a server in an information-theoretic setting require quantum communication. In this work, we investigate methods to reduce the communication overhead. First, we establish an impossibility result by proving that local processes on the server side cannot increase the number of qubits required for the computation. We develop a series of no-go results that prohibit such a process within an information-theoretic framework. Second, we present a possibility result by introducing the notion of selectively blind quantum computing (SBQC), a protocol that minimizes the number of encrypted qubits in the computation when delegating one computation from a pre-known set of computations. This approach, which we term can reduce communication costs drastically depending on the type of the possible computations and the differences between them.

Authentication is a fundamental building block of secure quantum networks, essential for quantum cryptographic protocols and often debated as a key limitation of quantum key distribution (QKD) in security standards. Most quantum-safe authentication schemes rely on small pre-shared keys or post-quantum computational assumptions. In this work, we introduce a new authentication approach that combines hardware assumptions, particularly Physical Unclonable Functions (PUFs), along with fundamental quantum properties of non-local states, such as local indistinguishability, to achieve a provable security in an entanglement-based protocol. We propose two protocols for different scenarios in entanglement-enabled quantum networks. The first protocol, referred to as the offline protocol, requires pre-distributed entangled states but no quantum communication during the process of authentication. It enables a server to authenticate clients at any time with only minimal classical communication. The second, an online protocol, requires quantum communication but only necessitates entangled state generation on the Prover’s side. For this, we introduce a novel hardware module, the Hybrid Entangled PUF (HEPUF). Both protocols use weakly secure, off-the-shelf classical PUFs as their hardware module, yet we prove that quantum properties such as local indistinguishability enable exponential security for authentication, even in a single round. We provide a full security analysis for both protocols and establish them as the first entanglement-based extension of hardware-based quantum authentication. These protocols are suitable for implementation across various platforms, particularly photonics-based ones, and offer a practical and flexible solution to the long-standing challenge of authentication in quantum communication networks.

We show the security of multi-user key establishment on a single line of quantum communication. More precisely, we consider a quantum communication architecture where the qubit generation and measurement happen at the two ends of the line, whilst intermediate parties are limited to single-qubit unitary transforms. This network topology has been previously introduced to implement quantum-assisted secret-sharing protocols for classical data, as well as the key establishment, and secure computing. This architecture has numerous advantages. The intermediate nodes are only using simplified hardware, which makes them easier to implement. Moreover, key establishment between arbitrary pairs of parties in the network does not require key routing through intermediate nodes. This is in contrast with quantum key distribution networks for which non- adjacent nodes need intermediate ones to route keys, thereby revealing these keys to intermediate parties and consuming previously established ones to secure the routing process. Our main result is to show the security of key establishment on quantum line networks. We show the security using the framework of abstract cryptography. This immediately makes the security composable, showing that the keys can be used for encryption or other tasks.

In this paper, we continue the line of work initiated by Boneh and Zhandry at CRYPTO 2013 and EUROCRYPT 2013 in which they formally define the notion of unforgeability against quantum adversaries. We develop a general and parameterised quantum game-based security model unifying unforgeability for both classical and quantum constructions allowing us for the first time to present a complete quantum cryptanalysis framework for unforgeability. In particular, we prove how our definitions subsume previous ones while considering more fine-grained adversarial models, capturing the full spectrum of superposition attacks. The subtlety here resides in the characterisation of a forgery. We show that the strongest level of unforgeability in our framework, namely existential unforgeability, can only be achieved if only orthogonal to previously queried messages are considered to be forgeries. We further show that deterministic constructions can only achieve the weaker notion of unforgeability, that is selective unforgeability, against such adversaries, but that selective unforgeability breaks if more general quantum adversaries (capable of general superposition attacks) are considered. On the other hand, we show that PRF is sufficient for constructing a selective unforgeable classical primitive against full quantum adversaries. Moreover, we show similar positive results relying on Pseudorandom Unitaries (PRU) for quantum primitives. \\ These results demonstrate the generality of our framework that could be applicable to other primitives beyond the cases analysed in this paper.

Cryptanalysis of quantum cryptographic systems generally involves finding optimal adversarial attack strategies on the underlying protocols. The core principle of modeling quantum attacks often reduces to the ability of the adversary to clone unknown quantum states and to extract thereby meaningful secret information. Explicit optimal attack strategies typically require high computational resources due to large circuit depths or, in many cases, are unknown. Here we introduce variational quantum cloning (VarQlone), a cryptanalysis algorithm based on quantum machine learning, which allows an adversary to obtain optimal approximate cloning strategies with short depth quantum circuits, trained using hybrid classical-quantum techniques. The algorithm contains operationally meaningful cost functions with theoretical guarantees, quantum circuit structure learning and gradient-descent-based optimization. Our approach enables the end-to-end discovery of hardware-efficient quantum circuits to clone specific families of quantum states, which we demonstrate in implementation on the Rigetti Aspen quantum hardware. We connect these results to quantum cryptographic primitives and derive explicit attacks facilitated by VarQlone. We expect that quantum machine learning will serve as a resource for improving attacks on current and future quantum cryptographic protocols.

Recently, major progress has been made towards the realisation of the quantum internet to enable a broad range of applications that would be out of reach for classical internet. Most of these applications such as delegated quantum computation require running a secure identification protocol between a low-resource and a high-resource party to provide secure communication. Physical Unclonable Functions (PUFs) have been shown as resource-efficient hardware solutions for providing secure identification schemes in both classical and quantum settings. In this work, we propose two identification protocols based on quantum PUFs (qPUFs) as defined recently by Arapinis et al. In the first protocol, the low-resource party wishes to prove its identity to the high-resource party and in the second protocol, it is vice versa. Unlike existing identification protocols based on Quantum Read-out of PUFs which rely on the security against a specific family of attacks, our protocols provide provable exponential security against any Quantum Polynomial-Time adversary with only polynomial resource parties. We provide a comprehensive comparison between the two proposed protocols in terms of resources such as quantum memory and computing ability required in both parties as well as the communication overhead between them. A stand-out feature of our second protocol is secure identification of a high-resource party by running a purely classical verification algorithm. This is achieved by delegating quantum operations to the high-resource party and utilising the resulting classical outcomes for identification. An interesting application idea that emerges from our second protocol is certification or benchmarking of general quantum computation schemes based on purely running a classical test on the resulting measurement outcomes.