Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

APL Photonics

Unclonable photonic keys hardened against machine learning attacks

Bryan T. Bosworth, Iskandar Atakhodjaev, Michael R. Kossey, Brian C. Grubel, Daniel S. Vresilovic, Jasper R. Stroud, Neil MacFarlane, Jesús Villalba, Najim Dehak, A. Brinton Cooper, Mark A. Foster, Amy C. Foster

Published Jan 1, 2020• 39 citations
Publisher

Abstract

The hallmark of the information age is the ease with which information is stored, accessed, and shared throughout the globe. This is enabled, in large part, by the simplicity of duplicating digital information without error. Unfortunately, an ever-growing consequence is the global threat to security and privacy enabled by our digital reliance. Specifically, modern secure communications and authentication suffer from formidable threats arising from the potential for copying of secret keys stored in digital media. With relatively little transfer of information, an attacker can impersonate a legitimate user, publish malicious software that is automatically accepted as safe by millions of computers, or eavesdrop on countless digital exchanges. To address this vulnerability, a new class of cryptographic devices known as physical unclonable functions (PUFs) are being developed. PUFs are modern realizations of an ancient concept, the physical key, and offer an attractive alternative for digital key storage. A user derives a digital key from the PUF’s physical behavior, which is sensitive to physical idiosyncrasies that are beyond fabrication tolerances. Thus, unlike conventional physical keys, a PUF cannot be duplicated and only the holder can extract the digital key. However, emerging machine learning (ML) methods are remarkably adept at learning behavior via training, and if such algorithms can learn to emulate a PUF, then the security is compromised. Unfortunately, such attacks are highly successful against conventional electronic PUFs. Here, we investigate ML attacks against a nonlinear silicon photonic PUF, a novel design that leverages nonlinear optical interactions in chaotic silicon microcavities. First, we investigate these devices’ resistance to cloning during fabrication and demonstrate their use as a source of large volumes of cryptographic key material. Next, we demonstrate that silicon photonic PUFs exhibit resistance to state-of-the-art ML attacks due to their nonlinearity and finally validate this resistance in an encryption scenario.

DOI
10.1063/1.5100178

Related papers

Clones of the Unclonable: Nanoduplicating Optical PUFs and Applications

Evangelos Marakis, Ulrich Rührmair, M. Lachner +3 more

Physical unclonable functions (PUFs), physical objects that are practically unclonable because of their andom and uncontrollable manufacturing variations, are becoming increasingly popular as security primitives and unique identifiers in a fully digitized world. One of the central PUF premises states that both friends and foes, both legitimate manufacturers and external attackers alike, cannot clone a PUF, producing two instances that are the same. Using the latest nanofabrication techniques, we show that this premise is not always met: We demonstrate the possibility of effective PUF duplication through sophisticated manufacturers by producing 63 copies of a non-trivial optical scattering structure which exhibit essentially the same scattering behavior. The remaining minuscule differences are close to or below noise levels, whence the duplicates have to be considered fully equivalent from a PUF perspective. The possibility for manufacturer-based optical PUF duplication has positive and negative consequences at the same time: While fully breaking the security of certain schemes, it enables new applications, too. For example, it facilitates unforgeable labels for valuable items; the first key-free group identification schemes over digital networks; or new types of encryption/decryption devices that do not contain secret keys.

arXiv:2212.12495•Dec 23, 2022•9 citations

Scalable and CMOS compatible silicon photonic physical unclonable functions for supply chain assurance

Farhan Bin Tarik, Azadeh Famili, Yingjie Lao +1 more

Abstract We demonstrate the uniqueness, unclonability and secure authentication of N = 56 physical unclonable functions (PUFs) realized from silicon photonic moiré quasicrystal interferometers. Compared to prior photonic-PUF demonstrations typically limited in scale to only a handful of unique devices and on the order of 10 false authentication attempts, this work examines > 103 inter-device comparisons and false authentication attempts. Device fabrication is divided across two separate fabrication facilities, allowing for cross-fab analysis and emulation of a malicious foundry with exact knowledge of the PUF photonic circuit design and process. Our analysis also compares cross-correlation based authentication to the traditional Hamming distance method and experimentally demonstrates an authentication error rate AER = 0%, false authentication rate FAR = 0%, and an estimated probability of cloning below 10−30. This work validates the potential scalability of integrated photonic-PUFs which can attractively leverage mature wafer-scale manufacturing and automated contact-free optical probing. Such structures show promise for authenticating hardware in the untrusted supply chain or augmenting conventional electronic-PUFs to enhance system security.

Scientific Reports•Sep 19, 2022•33 citations

Quantum-Secure Physical Unclonable Function enabled by Silicon Photonics Integrated Circuits

G. Sarantoglou, N. Tzekas, G. Moustakas +8 more

Physical Unclonable Functions (PUFs) are hardware security primitives whose inherent physical complexity can be exploited for secure authentication and cryptographic key generation. Silicon photonic devices, owing to their suitability for quantum and artificial intelligence applications alongside standard CMOS fabrication processes, constitute a highly promising substrate for integrated multifunctional PUFs. Despite the advanced security guarantees offered by quantum cryptographic protocols and the central role of silicon photonics in quantum technologies, quantum readout strategies based on single-photon states for photonic PUFs remain largely unexplored. In this work, we experimentally demonstrate a silicon nitride (SiN) programmable photonic Mach Zehnder interferometer mesh that implements a unitary transformation and operates as a PUF, whose secret physical signature arises from uncontrollable waveguide variations during fabrication. Using experimentally derived parameters from the SiN integrated mesh, we further introduce and numerically evaluate a quantum readout protocol that combines single-photon states with PUFs. Maximally mixed quantum states are employed to conceal the underlying unitary transformation from passive eavesdropping. Security against adversaries possessing devices fabricated under similar conditions is assessed, with authentication performance quantified through Monte Carlo analysis of the false acceptance and false rejection rates as a function of the number of detected events and corrected errors. The results indicate exceptional performance with equal error rates as low as 10 to the minus 14, highlighting the potential of quantum secure PUFs for high security authentication applications.

physics.opticsCross-listedarXiv:2605.14959•May 15, 2026•1 citation

Integrated photonic cryptographic key generator using low-power MEMS-on-PIC technology

Martin Blasl, Meysam Namdari, Maximilian Wagner +1 more

To improve data security and authentication, we designed and fabricated a cryptographic key generator based on a photonic integrated circuit that converts a digital input key into a digital output key by means of a physical unclonable function (PUF). The PUF is realized by an imperfect multi-mode interferometer controlled by low-power micro-electromechanical system (MEMS) phase shifters. An analytical model was derived and used to prove the randomness of the generated digital keys, and the model was validated by measurements of a first proof-of-concept demonstrator. The demonstrator was fabricated using a silicon nitride photonic integrated circuit (PIC) platform and our recently developed MEMS-on-PIC technology.

Journal of Optical Microsystems•Aug 19, 2024•4 citations

Quantum readout of Physical Unclonable Functions

Boris Škorić

Physical unclonable functions (PUFs) are physical structures that are hard to clone and have a unique challenge-response behavior. The term PUF was coined by Pappu et al. in 2001. That work triggered a lot of interest, and since then a substantial number of papers has been written about the use of a wide variety of physical structures for different security purposes such as identification, authentication, read-proof key storage, key distribution, tamper evidence, anti-counterfeiting, software-to-hardware binding and trusted computing. In this paper we propose a new security primitive: the quantum-readout PUF (QR-PUF). This is a classical PUF, without internal quantum degrees of freedom, which is challenged using a quantum state, e.g. a single-photon state, and whose response is also a quantum state. By the no-cloning property of unknown quantum states, attackers cannot intercept challenges or responses without noticeably disturbing the readout process. Thus, a verifier who sends quantum states as challenges and receives the correct quantum states back can be certain that he is probing a specific QR-PUF without disturbances, even if the QR-PUF is far away "in the field" and under hostile control. For PUFs whose information content is not exceedingly large, all currently known PUF-based authentication and anti-counterfeiting schemes require trusted readout devices in the field. Our quantum readout scheme has no such requirement. Furthermore, we show how the QR-PUF authentication scheme can be interwoven with quantum key exchange (QKE), leading to an authenticated QKE protocol between two parties. This protocol has the special property that it requires no a priori secret shared by the two parties, and that the quantum channel is the authenticated channel, allowing for an unauthenticated classical channel. We provide security proofs for a limited class of attacks. The proofs depend on the physical unclonability of PUFs and on the practical infeasibility of building a quantum computer.

International Journal of Quantum Information•Feb 1, 2012•63 citations

Quantum PUF and its Applications with Information Theoretic Analysis

Kumar Nilesh, Christian Deppe, Holger Boche

Physical Unclonable Functions (PUFs) have garnered significant attention for their potential in enhancing the security of IoT devices, offering an alternative to traditional computational-based approaches. Despite their potential, classical PUFs suffer from critical drawbacks, including vulnerability to machine learning attacks and inadequate defense against quantum adversaries. To address these limitations, alternative architectures utilizing quantum mechanical principles, known as Quantum PUFs (QPUFs), have been proposed. This paper presents a comprehensive survey of the diverse range of QPUFs, detailing their architectures and security features. Additionally, it explores the information-theoretic applications of QPUFs in IoT, including secret key generation, secure storage, authentication, and identification. The findings indicate that classical results translate effectively to the quantum domain in most cases, with QPUFs exhibiting larger capacity regions compared to their classical counterparts.

IEEE World Forum on Internet of Things (WF-IoT)•Nov 10, 2024•12 citations
The hallmark of the information age is the ease with which information is stored, accessed, and shared throughout the globe. This is enabled, in large part, by the simplicity of duplicating digital information without error. Unfortunately, an ever-growing consequence is the global threat to security and privacy enabled by our digital reliance. Specifically, modern secure communications and authentication suffer from formidable threats arising from the potential for copying of secret keys stored in digital media. With relatively little transfer of information, an attacker can impersonate a legitimate user, publish malicious software that is automatically accepted as safe by millions of computers, or eavesdrop on countless digital exchanges. To address this vulnerability, a new class of cryptographic devices known as physical unclonable functions (PUFs) are being developed. PUFs are modern realizations of an ancient concept, the physical key, and offer an attractive alternative for digital key storage. A user derives a digital key from the PUF’s physical behavior, which is sensitive to physical idiosyncrasies that are beyond fabrication tolerances. Thus, unlike conventional physical keys, a PUF cannot be duplicated and only the holder can extract the digital key. However, emerging machine learning (ML) methods are remarkably adept at learning behavior via training, and if such algorithms can learn to emulate a PUF, then the security is compromised. Unfortunately, such attacks are highly successful against conventional electronic PUFs. Here, we investigate ML attacks against a nonlinear silicon photonic PUF, a novel design that leverages nonlinear optical interactions in chaotic silicon microcavities. First, we investigate these devices’ resistance to cloning during fabrication and demonstrate their use as a source of large volumes of cryptographic key material. Next, we demonstrate that silicon photonic PUFs exhibit resistance to state-of-the-art ML attacks due to their nonlinearity and finally validate this resistance in an encryption scenario.
Physical unclonable functions (PUFs), physical objects that are practically unclonable because of their andom and uncontrollable manufacturing variations, are becoming increasingly popular as security primitives and unique identifiers in a fully digitized world. One of the central PUF premises states that both friends and foes, both legitimate manufacturers and external attackers alike, cannot clone a PUF, producing two instances that are the same. Using the latest nanofabrication techniques, we show that this premise is not always met: We demonstrate the possibility of effective PUF duplication through sophisticated manufacturers by producing 63 copies of a non-trivial optical scattering structure which exhibit essentially the same scattering behavior. The remaining minuscule differences are close to or below noise levels, whence the duplicates have to be considered fully equivalent from a PUF perspective. The possibility for manufacturer-based optical PUF duplication has positive and negative consequences at the same time: While fully breaking the security of certain schemes, it enables new applications, too. For example, it facilitates unforgeable labels for valuable items; the first key-free group identification schemes over digital networks; or new types of encryption/decryption devices that do not contain secret keys.
Abstract We demonstrate the uniqueness, unclonability and secure authentication of N = 56 physical unclonable functions (PUFs) realized from silicon photonic moiré quasicrystal interferometers. Compared to prior photonic-PUF demonstrations typically limited in scale to only a handful of unique devices and on the order of 10 false authentication attempts, this work examines > 103 inter-device comparisons and false authentication attempts. Device fabrication is divided across two separate fabrication facilities, allowing for cross-fab analysis and emulation of a malicious foundry with exact knowledge of the PUF photonic circuit design and process. Our analysis also compares cross-correlation based authentication to the traditional Hamming distance method and experimentally demonstrates an authentication error rate AER = 0%, false authentication rate FAR = 0%, and an estimated probability of cloning below 10−30. This work validates the potential scalability of integrated photonic-PUFs which can attractively leverage mature wafer-scale manufacturing and automated contact-free optical probing. Such structures show promise for authenticating hardware in the untrusted supply chain or augmenting conventional electronic-PUFs to enhance system security.
Physical Unclonable Functions (PUFs) are hardware security primitives whose inherent physical complexity can be exploited for secure authentication and cryptographic key generation. Silicon photonic devices, owing to their suitability for quantum and artificial intelligence applications alongside standard CMOS fabrication processes, constitute a highly promising substrate for integrated multifunctional PUFs. Despite the advanced security guarantees offered by quantum cryptographic protocols and the central role of silicon photonics in quantum technologies, quantum readout strategies based on single-photon states for photonic PUFs remain largely unexplored. In this work, we experimentally demonstrate a silicon nitride (SiN) programmable photonic Mach Zehnder interferometer mesh that implements a unitary transformation and operates as a PUF, whose secret physical signature arises from uncontrollable waveguide variations during fabrication. Using experimentally derived parameters from the SiN integrated mesh, we further introduce and numerically evaluate a quantum readout protocol that combines single-photon states with PUFs. Maximally mixed quantum states are employed to conceal the underlying unitary transformation from passive eavesdropping. Security against adversaries possessing devices fabricated under similar conditions is assessed, with authentication performance quantified through Monte Carlo analysis of the false acceptance and false rejection rates as a function of the number of detected events and corrected errors. The results indicate exceptional performance with equal error rates as low as 10 to the minus 14, highlighting the potential of quantum secure PUFs for high security authentication applications.
To improve data security and authentication, we designed and fabricated a cryptographic key generator based on a photonic integrated circuit that converts a digital input key into a digital output key by means of a physical unclonable function (PUF). The PUF is realized by an imperfect multi-mode interferometer controlled by low-power micro-electromechanical system (MEMS) phase shifters. An analytical model was derived and used to prove the randomness of the generated digital keys, and the model was validated by measurements of a first proof-of-concept demonstrator. The demonstrator was fabricated using a silicon nitride photonic integrated circuit (PIC) platform and our recently developed MEMS-on-PIC technology.
Physical unclonable functions (PUFs) are physical structures that are hard to clone and have a unique challenge-response behavior. The term PUF was coined by Pappu et al. in 2001. That work triggered a lot of interest, and since then a substantial number of papers has been written about the use of a wide variety of physical structures for different security purposes such as identification, authentication, read-proof key storage, key distribution, tamper evidence, anti-counterfeiting, software-to-hardware binding and trusted computing. In this paper we propose a new security primitive: the quantum-readout PUF (QR-PUF). This is a classical PUF, without internal quantum degrees of freedom, which is challenged using a quantum state, e.g. a single-photon state, and whose response is also a quantum state. By the no-cloning property of unknown quantum states, attackers cannot intercept challenges or responses without noticeably disturbing the readout process. Thus, a verifier who sends quantum states as challenges and receives the correct quantum states back can be certain that he is probing a specific QR-PUF without disturbances, even if the QR-PUF is far away "in the field" and under hostile control. For PUFs whose information content is not exceedingly large, all currently known PUF-based authentication and anti-counterfeiting schemes require trusted readout devices in the field. Our quantum readout scheme has no such requirement. Furthermore, we show how the QR-PUF authentication scheme can be interwoven with quantum key exchange (QKE), leading to an authenticated QKE protocol between two parties. This protocol has the special property that it requires no a priori secret shared by the two parties, and that the quantum channel is the authenticated channel, allowing for an unauthenticated classical channel. We provide security proofs for a limited class of attacks. The proofs depend on the physical unclonability of PUFs and on the practical infeasibility of building a quantum computer.
Physical Unclonable Functions (PUFs) have garnered significant attention for their potential in enhancing the security of IoT devices, offering an alternative to traditional computational-based approaches. Despite their potential, classical PUFs suffer from critical drawbacks, including vulnerability to machine learning attacks and inadequate defense against quantum adversaries. To address these limitations, alternative architectures utilizing quantum mechanical principles, known as Quantum PUFs (QPUFs), have been proposed. This paper presents a comprehensive survey of the diverse range of QPUFs, detailing their architectures and security features. Additionally, it explores the information-theoretic applications of QPUFs in IoT, including secret key generation, secure storage, authentication, and identification. The findings indicate that classical results translate effectively to the quantum domain in most cases, with QPUFs exhibiting larger capacity regions compared to their classical counterparts.