Damien Rontani
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Damien Rontani has a total of 54 co-authors. Here are the top co-authors by number of shared publications:
- Piotr Antonik
CentraleSupélec
7 shared publications - Nicolas Marsal
Université de Lorraine, CentraleSupélec, LMOPS, Metz, France
4 shared publications - Alexandre Locquet
School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, USA
3 shared publications - David S. Citrin
Georgia Tech-CNRS IRL2958, Georgia Tech-Europe, Metz, France
3 shared publications - Julien Rauch
Universit Paris-Saclay,LISN UMR-9015, CNRS,Orsay,France,F-91405
3 shared publications Nicolas Marsal
LMOPS EA 4423 Laboratory CentraleSup élec & Universit é de Lorraine F-57070 Metz France
3 shared publications
Latest Publications
- Open Access
- 1 October 2026
- Published by Springer Nature in Scientific Reports
Deep binarized photonic reservoir computing for ultrafast multimedia signal processing
We present a deep photonic neural network architecture based on ultrafast binary optical modulation using a digital micromirror device (DMD), optical scattering in a random medium, high-speed photodetection with a CMOS sensor, and a time-multiplexed multilayer structure. Operating at Gigabit-per-second (Gb/s) optical information throughput, our system based on the reservoir computing (RC) framework achieves state-of-the-art performance across various multimedia tasks, including video, image and speech recognition. We show that the careful optimization of key physical intra- and inter-layer hyperparameters can significantly enhance the deep photonic RC system’s ability to extract relevant temporal and spatial features by balancing memory retention and dynamical response of individual layers. This approach paves the way for highly scalable hierarchical photonic reservoir computing systems for high-throughput real-time multimedia signal processing.
- 1 October 2026
- Published by Springer Nature in Scientific Reports
Open AccessDeep binarized photonic reservoir computing for ultrafast multimedia signal processing
We present a deep photonic neural network architecture based on ultrafast binary optical modulation using a digital micromirror device (DMD), optical scattering in a random medium, high-speed photodetection with a CMOS sensor, and a time-multiplexed multilayer structure. Operating at Gigabit-per-second (Gb/s) optical information throughput, our system based on the reservoir computing (RC) framework achieves state-of-the-art performance across various multimedia tasks, including video, image and speech recognition. We show that the careful optimization of key physical intra- and inter-layer hyperparameters can significantly enhance the deep photonic RC system’s ability to extract relevant temporal and spatial features by balancing memory retention and dynamical response of individual layers. This approach paves the way for highly scalable hierarchical photonic reservoir computing systems for high-throughput real-time multimedia signal processing.
- 30 March 2026
- Published by Agence Bibliographique de l'Enseignement Supérieur
Nonlinear dynamics of photonic components. Chaos cryptography and multiplexing
Dynamique non-linéaire de composants photoniques. Cryptographie par chaos et multiplexage: L’application du concept de synchronisation appliqué aux composants photoniques non-linéaires a permis l’avènement des communications chaotiques optiques. Les systèmes optoélectronique dans ces chaines de transmission ont déjà démontré leur potentiel en termes de sécurité additionnelle au niveau de la couche physique des réseaux optiques. Cependant, la quantification du niveau de sécurité et l’absence d’un cadre déduit aux aspects multi-utilisateurs (techniques de multiplexage spécifiques) ont fortement freiné le déploiement de ces techniques à large échelle. La recherche effectuée dans le cadre de cette thèse a contribué à l’avancement de ces deux questions ouvertes. Dans premier temps, nous nous sommes intéressés à la sécurité d’une classe de générateur optique particulière: les lasers à semi-conducteur à cavité externe (ECSL). Nous proposons d’attaquer le système dans le contexte le plus difficile, celui dans lequel aucune information n’est a priori disponible. La sortie du laser chaotique (l’intensité optique) est enregistrée, et les séries temporelles obtenues sont analysées. La sécurité est comprise dans ce contexte comme étant la quantité d’information sur les paramètres du système (analogue d’une clé dans les systèmes de cryptage conventionnels) et/ou la fonction non-linéaire du système (l’équivalent de l’algorithme de cryptage). Un ECSL est un système possédant un délai (aussi appelé retard), un paramètre critique pour la sécurité. Nous avons étudié l’influence des paramètres ajustable de l’ECSL chaotique sur l’identification du délai: l’intensité de la rétroaction optique, la valeur du délai, et le courant alimentant le laser (aussi appelé courant de pompe). Dans un deuxième temps nous interprétons ces résultats sur la base des régimes dynamiques précédent l’apparition du chaos dans l’ECSL. Nous avons proposé par la suite une architecture pour multiplexer des signaux chaotiques optiques générés par des ECSL. Nous démontrons la supériorité de cette approche en terme d’efficacité spectrale relativement aux méthodes de multiplexage en longueur d’onde (wavelength division multiplexing, WDM) appliquées aux communications optiques par chaos (aussi connues sous le nom de chaotic WDM ). Nous avons adapté un concept fondamental de la théorie de la synchronisation: la décomposition active-passive (APD) en utilisant des composants optiques simples. Nous démontrons la possibilité de multiplexer et démultiplexage de deux signaux chaotiques optiques par synchronisation (en utilisant deux émetteurs et deux récepteurs). Les performances et la robustesse de cette structure sont analysées ainsi que la possibilité de crypter et de décrypter des messages. Après cela, nous avons utilisé une classe de systèmes optoélectroniques différente de celle correspondant au deux premières sections, avec l’objectif d’utiliser un seul oscillateur chaotique pour encoder plusieurs messages au lieu de considérer un oscillateur par message. A cette fin, nous avons modifié une structure d’un générateur de chaos électro-optique existant dans la littérature en lui ajoutant plusieurs boucles de rétroactions non-linéaires. Chacune d’elle est utilisée pour l’encryptage d’un message, réalisée, par exemple, par la modulation du gain non-linéaire de la boucle. Nous avons analysé différentes configurations possibles, ainsi que les propriétés des signaux chaotiques générés au sein de chaque boucle et utilisés pour transporter les différents messages. Nous avons expliqué dans quelle mesure l’orthogonalité (ou décorrélation) entre les différents signaux peut être utilisée à notre avantage pour dériver des équations de décodage de faible complexité algorithmique (comme fonction du nombre d’utilisateurs). Enfin, nous analysons comment certains phénomènes d’interférences entre signaux porteurs peuvent être pris en compte afin de toujours permettre la récupération des messages.
- 30 March 2026
- Published by Agence Bibliographique de l'Enseignement Supérieur
Nonlinear dynamics of photonic components. Chaos cryptography and multiplexing
Dynamique non-linéaire de composants photoniques. Cryptographie par chaos et multiplexage: L’application du concept de synchronisation appliqué aux composants photoniques non-linéaires a permis l’avènement des communications chaotiques optiques. Les systèmes optoélectronique dans ces chaines de transmission ont déjà démontré leur potentiel en termes de sécurité additionnelle au niveau de la couche physique des réseaux optiques. Cependant, la quantification du niveau de sécurité et l’absence d’un cadre déduit aux aspects multi-utilisateurs (techniques de multiplexage spécifiques) ont fortement freiné le déploiement de ces techniques à large échelle. La recherche effectuée dans le cadre de cette thèse a contribué à l’avancement de ces deux questions ouvertes. Dans premier temps, nous nous sommes intéressés à la sécurité d’une classe de générateur optique particulière: les lasers à semi-conducteur à cavité externe (ECSL). Nous proposons d’attaquer le système dans le contexte le plus difficile, celui dans lequel aucune information n’est a priori disponible. La sortie du laser chaotique (l’intensité optique) est enregistrée, et les séries temporelles obtenues sont analysées. La sécurité est comprise dans ce contexte comme étant la quantité d’information sur les paramètres du système (analogue d’une clé dans les systèmes de cryptage conventionnels) et/ou la fonction non-linéaire du système (l’équivalent de l’algorithme de cryptage). Un ECSL est un système possédant un délai (aussi appelé retard), un paramètre critique pour la sécurité. Nous avons étudié l’influence des paramètres ajustable de l’ECSL chaotique sur l’identification du délai: l’intensité de la rétroaction optique, la valeur du délai, et le courant alimentant le laser (aussi appelé courant de pompe). Dans un deuxième temps nous interprétons ces résultats sur la base des régimes dynamiques précédent l’apparition du chaos dans l’ECSL. Nous avons proposé par la suite une architecture pour multiplexer des signaux chaotiques optiques générés par des ECSL. Nous démontrons la supériorité de cette approche en terme d’efficacité spectrale relativement aux méthodes de multiplexage en longueur d’onde (wavelength division multiplexing, WDM) appliquées aux communications optiques par chaos (aussi connues sous le nom de chaotic WDM ). Nous avons adapté un concept fondamental de la théorie de la synchronisation: la décomposition active-passive (APD) en utilisant des composants optiques simples. Nous démontrons la possibilité de multiplexer et démultiplexage de deux signaux chaotiques optiques par synchronisation (en utilisant deux émetteurs et deux récepteurs). Les performances et la robustesse de cette structure sont analysées ainsi que la possibilité de crypter et de décrypter des messages. Après cela, nous avons utilisé une classe de systèmes optoélectroniques différente de celle correspondant au deux premières sections, avec l’objectif d’utiliser un seul oscillateur chaotique pour encoder plusieurs messages au lieu de considérer un oscillateur par message. A cette fin, nous avons modifié une structure d’un générateur de chaos électro-optique existant dans la littérature en lui ajoutant plusieurs boucles de rétroactions non-linéaires. Chacune d’elle est utilisée pour l’encryptage d’un message, réalisée, par exemple, par la modulation du gain non-linéaire de la boucle. Nous avons analysé différentes configurations possibles, ainsi que les propriétés des signaux chaotiques générés au sein de chaque boucle et utilisés pour transporter les différents messages. Nous avons expliqué dans quelle mesure l’orthogonalité (ou décorrélation) entre les différents signaux peut être utilisée à notre avantage pour dériver des équations de décodage de faible complexité algorithmique (comme fonction du nombre d’utilisateurs). Enfin, nous analysons comment certains phénomènes d’interférences entre signaux porteurs peuvent être pris en compte afin de toujours permettre la récupération des messages.
- 5 February 2026
- Published by Springer Nature
- p. 243-251
Towards a Quantum Generative Graph-Based Clustering for Molecule Discovery
- 5 February 2026
- Published by Springer Nature
- p. 243-251
Towards a Quantum Generative Graph-Based Clustering for Molecule Discovery
- Open Access
- 17 December 2025
- Published by IOP Publishing in Journal of Physics: Photonics
- Vol. 8 (2)
Photonic reservoir computing: A thematic review
Reservoir computing offers a versatile approach to temporal information processing by exploiting the dynamics of nonlinear systems to enable learning with a linear readout. Photonic hardware is particularly attractive for this paradigm, thanks to ultrafast dynamics and massive parallelism across multiple accessible degrees of freedom of light. In this thematic review, we chart the development of photonic reservoir computing across hardware platforms, from delay-based architectures and integrated photonic circuit implementations to free-space systems. We also discuss how the input and output layers influence the performance of photonic reservoirs. We present the explored learning algorithms that go beyond linear regression, and list some popular benchmark tasks and shed light on some key applications that have been investigated in the literature. Finally, we discuss emerging unconventional implementations. The overall goal of this review is to provide a panoramic overview of photonic reservoir computing, useful for both practitioners and newcomers to the field.
- 17 December 2025
- Published by IOP Publishing in Journal of Physics: Photonics
- Vol. 8 (2)
Open AccessPhotonic reservoir computing: A thematic review
Reservoir computing offers a versatile approach to temporal information processing by exploiting the dynamics of nonlinear systems to enable learning with a linear readout. Photonic hardware is particularly attractive for this paradigm, thanks to ultrafast dynamics and massive parallelism across multiple accessible degrees of freedom of light. In this thematic review, we chart the development of photonic reservoir computing across hardware platforms, from delay-based architectures and integrated photonic circuit implementations to free-space systems. We also discuss how the input and output layers influence the performance of photonic reservoirs. We present the explored learning algorithms that go beyond linear regression, and list some popular benchmark tasks and shed light on some key applications that have been investigated in the literature. Finally, we discuss emerging unconventional implementations. The overall goal of this review is to provide a panoramic overview of photonic reservoir computing, useful for both practitioners and newcomers to the field.
- Open Access
- 21 March 2025
- Published by Springer Nature in Light: Science & Applications
- Vol. 14 (1) , 1-9
Optical chaotic signal recovery in turbulent environments using a programmable optical processor
Optical chaos offers a promising approach to establishing secure communication at high data rates in a shared physical channel, like optical fibers and free space. However, the required synchronization between the transmitter and the receiver can be severely impaired by the nonidealities of the optical link. In particular, free-space optical communications are affected by atmospheric turbulence, which causes beam scintillation and results in time-varying fading of the optical intensity at the receiver side. In this work, we investigate experimentally the propagation of a chaotic signal in an indoor optical link with controllable synthetic turbulence, and we show that the degradation of chaos properties caused by the turbulent environment can be fully mitigated in the optical domain using an adaptive multi-aperture receiver. The proposed receiver integrates a two-dimensional array of optical antennas and a programmable optical processor (POP) on a silicon photonic platform. With respect to a conventional single-aperture receiver, the POP-assisted receiver recovers the complex dynamics of the optical chaos, ensuring a high degree of correlation between the transmitted signal and the received signal, even for a high degree of turbulence. Our results demonstrate the possibility of establishing and maintaining reliable, secure communication in a chaos-based crypto-system in a free space optical link of km-range length.
- 21 March 2025
- Published by Springer Nature in Light: Science & Applications
- Vol. 14 (1) , 1-9
Open AccessOptical chaotic signal recovery in turbulent environments using a programmable optical processor
Optical chaos offers a promising approach to establishing secure communication at high data rates in a shared physical channel, like optical fibers and free space. However, the required synchronization between the transmitter and the receiver can be severely impaired by the nonidealities of the optical link. In particular, free-space optical communications are affected by atmospheric turbulence, which causes beam scintillation and results in time-varying fading of the optical intensity at the receiver side. In this work, we investigate experimentally the propagation of a chaotic signal in an indoor optical link with controllable synthetic turbulence, and we show that the degradation of chaos properties caused by the turbulent environment can be fully mitigated in the optical domain using an adaptive multi-aperture receiver. The proposed receiver integrates a two-dimensional array of optical antennas and a programmable optical processor (POP) on a silicon photonic platform. With respect to a conventional single-aperture receiver, the POP-assisted receiver recovers the complex dynamics of the optical chaos, ensuring a high degree of correlation between the transmitted signal and the received signal, even for a high degree of turbulence. Our results demonstrate the possibility of establishing and maintaining reliable, secure communication in a chaos-based crypto-system in a free space optical link of km-range length.
- 15 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 02, 199-204
Investigating parallel execution of quantum Machine Learning circuits on superconducting hardware
Quantum Machine Learning algorithms generally rely on hybrid implementation with a classical learning loop running on a CPU and small quantum circuits running on a Quantum Processing Unit (QPU) for serialized processing batches of data. Considering the scarcity of quantum resources, parallelizing the execution of multiple instances of small quantum circuits instead of allocating a single circuit would better use quantum hardware resources. However, exploiting all available qubits in QPUs could result in a noisier operating condition and disrupt or slow the learning mechanism. This paper investigates the parallelization of QML algorithms on QPU for data clustering based on trainable generative models. We design a parallel macro-circuit with the Qiskit framework and measure their performance on an IBM superconducting quantum computer. A theoretical performance model is then proposed to determine the expected level of acceleration. Finally, we measure the impact of the QPU's occupancy on the loss functions of our QML algorithm, enabling us to identify the most reliable and exploitable acceleration range.
- 15 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 02, 199-204
Investigating parallel execution of quantum Machine Learning circuits on superconducting hardware
Quantum Machine Learning algorithms generally rely on hybrid implementation with a classical learning loop running on a CPU and small quantum circuits running on a Quantum Processing Unit (QPU) for serialized processing batches of data. Considering the scarcity of quantum resources, parallelizing the execution of multiple instances of small quantum circuits instead of allocating a single circuit would better use quantum hardware resources. However, exploiting all available qubits in QPUs could result in a noisier operating condition and disrupt or slow the learning mechanism. This paper investigates the parallelization of QML algorithms on QPU for data clustering based on trainable generative models. We design a parallel macro-circuit with the Qiskit framework and measure their performance on an IBM superconducting quantum computer. A theoretical performance model is then proposed to determine the expected level of acceleration. Finally, we measure the impact of the QPU's occupancy on the loss functions of our QML algorithm, enabling us to identify the most reliable and exploitable acceleration range.
- 1 August 2024
- Published by Springer Nature
- p. 282-297
Generative-Based Algorithm for Data Clustering on Hybrid Classical-Quantum NISQ Architecture
- 1 August 2024
- Published by Springer Nature
- p. 282-297
Generative-Based Algorithm for Data Clustering on Hybrid Classical-Quantum NISQ Architecture
- Open Access
- 1 November 2023
- Published by AIP Publishing in Chaos: An Interdisciplinary Journal of Nonlinear Science
- Vol. 33 (11)
Resonant properties of the memory capacity of a laser-based reservoir computer with filtered optoelectronic feedback
We provide a comprehensive analysis of the resonant properties of the memory capacity of a reservoir computer based on a semiconductor laser subjected to time-delayed filtered optoelectronic feedback. Our analysis reveals first how the memory capacity decreases sharply when the input-data clock cycle is slightly time-shifted from the time delay or its multiples. We attribute this effect to the inertial properties of the laser. We also report on the damping of the memory-capacity drop at resonance with a decrease of the virtual-node density and its broadening with the filtering properties of the optoelectronic feedback. These results are interpretated using the eigenspectrum of the reservoir obtained from a linear stability analysis. Then, we unveil an invariance in the minimum value of the memory capacity at resonance with respect to a variation of the number of nodes if the number is big enough and quantify how the filtering properties impact the system memory in and out of resonance.
- 1 November 2023
- Published by AIP Publishing in Chaos: An Interdisciplinary Journal of Nonlinear Science
- Vol. 33 (11)
Open AccessResonant properties of the memory capacity of a laser-based reservoir computer with filtered optoelectronic feedback
We provide a comprehensive analysis of the resonant properties of the memory capacity of a reservoir computer based on a semiconductor laser subjected to time-delayed filtered optoelectronic feedback. Our analysis reveals first how the memory capacity decreases sharply when the input-data clock cycle is slightly time-shifted from the time delay or its multiples. We attribute this effect to the inertial properties of the laser. We also report on the damping of the memory-capacity drop at resonance with a decrease of the virtual-node density and its broadening with the filtering properties of the optoelectronic feedback. These results are interpretated using the eigenspectrum of the reservoir obtained from a linear stability analysis. Then, we unveil an invariance in the minimum value of the memory capacity at resonance with respect to a variation of the number of nodes if the number is big enough and quantify how the filtering properties impact the system memory in and out of resonance.
- 11 May 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Terahertz Science and Technology
- Vol. 13 (4) , 396-404
THz-Based Monitoring of the Deformation of the Inner Woven Structure of Glass-Fiber Reinforced Polymers
This work exploits terahertz (THz) imaging in the real-space and the spatial-frequency domain to determine nondestructively the evolution of the main characteristics of a woven glass-fiber reinforced polymer (GFRP) laminate in three-dimensions in-situ during a tensile test. These characteristics include tow orientation and the weave period. The surface and the inner plies can be investigated with the proposed method. The results show how the tows tend to realign with the load direction. In addition, we detect a decrease in the weave period, manifesting a distortion of the surface and inner layers as the stress is increased. Contrary to conventional optical inspection, THz inspection can provide quantitative measurements in all material layers and not only on the surface. This technique could be helpful to monitor distortion and damage due to mechanical stress or defects incorporated during the manufacturing process and to validate parameters and results of draping process modeling, especially for the inner plies.
- 11 May 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Terahertz Science and Technology
- Vol. 13 (4) , 396-404
THz-Based Monitoring of the Deformation of the Inner Woven Structure of Glass-Fiber Reinforced Polymers
This work exploits terahertz (THz) imaging in the real-space and the spatial-frequency domain to determine nondestructively the evolution of the main characteristics of a woven glass-fiber reinforced polymer (GFRP) laminate in three-dimensions in-situ during a tensile test. These characteristics include tow orientation and the weave period. The surface and the inner plies can be investigated with the proposed method. The results show how the tows tend to realign with the load direction. In addition, we detect a decrease in the weave period, manifesting a distortion of the surface and inner layers as the stress is increased. Contrary to conventional optical inspection, THz inspection can provide quantitative measurements in all material layers and not only on the surface. This technique could be helpful to monitor distortion and damage due to mechanical stress or defects incorporated during the manufacturing process and to validate parameters and results of draping process modeling, especially for the inner plies.
- 1 January 2023
- Published by AIP Publishing in Chaos: An Interdisciplinary Journal of Nonlinear Science
- Vol. 33 (1) , 013116
Impact of filtering on photonic time-delay reservoir computing
We analyze the modification of the computational properties of a time-delay photonic reservoir computer with a change in its feedback bandwidth. For a reservoir computing configuration based on a semiconductor laser subject to filtered optoelectronic feedback, we demonstrate that bandwidth selection can lead to a flat-topped eigenvalue spectrum for which a large number of system frequencies are weakly damped as a result of the attenuation of modulational instability by feedback filtering. This spectral configuration allows for the optimization of the reservoir in terms of its memory capacity, while its computational ability appears to be only weakly affected by the characteristics of the filter.
- 1 January 2023
- Published by AIP Publishing in Chaos: An Interdisciplinary Journal of Nonlinear Science
- Vol. 33 (1) , 013116
Impact of filtering on photonic time-delay reservoir computing
We analyze the modification of the computational properties of a time-delay photonic reservoir computer with a change in its feedback bandwidth. For a reservoir computing configuration based on a semiconductor laser subject to filtered optoelectronic feedback, we demonstrate that bandwidth selection can lead to a flat-topped eigenvalue spectrum for which a large number of system frequencies are weakly damped as a result of the attenuation of modulational instability by feedback filtering. This spectral configuration allows for the optimization of the reservoir in terms of its memory capacity, while its computational ability appears to be only weakly affected by the characteristics of the filter.