José Azaña
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
José Azaña has a total of 198 co-authors. Here are the top co-authors by number of shared publications:
- Roberto Morandotti
Institut national de la recherche scientifique, Centre Énergie Matériaux Télécommunications, Varennes, Canada
20 shared publications - Luis Romero Cortés
Institut National de la Recherche Scientifique (INRS-EMT), Varennes, Canada
18 shared publications - David Moss
Swinburne University of Technology
15 shared publications - Benjamin Crockett
Institut National de la Recherche Scientifique – Énergie Matériaux et Télécommunications INRS-EMT, 1650 Lionel-Boulet Blvd., Varennes, QC J3X 1S2, Canada
14 shared publications - Sai Tak Chu
Department of Physics, City University of Hong Kong, Tat Chee Avenue, Hong Kong, China
14 shared publications - Reza Maram
Fonex Data Systems Inc., Montréal, Québec H4S 1P6, Canada
11 shared publications
Latest Publications
- Open Access
- 21 August 2026
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 12 (34)
Programmable photonic neural engine with all-optical nonlinear activation and 40,000 connections
The rapid advancement of artificial neural networks (ANNs) demands computational platforms with higher speed, energy efficiency, and scalability. Optical computing offers an appealing solution owing to the high-speed processing, inherent parallelism, and low energy consumption. However, existing optical neural network architectures face fundamental trade-offs between scalability, reconfigurability, and processing latency, constrained by the need to rely on electronic nonlinearities or fixed optical interconnections. Here, we report an end-to-end photonic neuromorphic engine that integrates all-optical nonlinearity into a loop-based, time-multiplexed photonic architecture. This approach enables all optical neurons to share common hardware, supporting enhanced connectivity and full reconfigurability with reduced latency. Implementations of four distinct network topologies, with a maximum of 40,000 optical connections for the single-layer perceptron, achieve digital-level inference accuracy and computational latency over two orders of magnitude shorter than state-of-the-art electronic processors, potentially paving the way for scalable, reconfigurable, and ultrafast optical artificial intelligence systems.
- 21 August 2026
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 12 (34)
Open AccessProgrammable photonic neural engine with all-optical nonlinear activation and 40,000 connections
The rapid advancement of artificial neural networks (ANNs) demands computational platforms with higher speed, energy efficiency, and scalability. Optical computing offers an appealing solution owing to the high-speed processing, inherent parallelism, and low energy consumption. However, existing optical neural network architectures face fundamental trade-offs between scalability, reconfigurability, and processing latency, constrained by the need to rely on electronic nonlinearities or fixed optical interconnections. Here, we report an end-to-end photonic neuromorphic engine that integrates all-optical nonlinearity into a loop-based, time-multiplexed photonic architecture. This approach enables all optical neurons to share common hardware, supporting enhanced connectivity and full reconfigurability with reduced latency. Implementations of four distinct network topologies, with a maximum of 40,000 optical connections for the single-layer perceptron, achieve digital-level inference accuracy and computational latency over two orders of magnitude shorter than state-of-the-art electronic processors, potentially paving the way for scalable, reconfigurable, and ultrafast optical artificial intelligence systems.
- Open Access
- 30 January 2026
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 12 (5) , eady8981
Quantum state revival via coherent energy redistribution
Processing and detecting quantum states with high fidelity are essential for enabling quantum advantages across many applications. However, these states are known to be fragile because of their sensitivity to losses, their inability to be amplified, and their susceptibility to decoherence. This makes them far more vulnerable to noise than classical signals, limiting their out-of-lab deployment. We demonstrate quantum coherent energy redistribution, which not only recovers entangled states buried in noise but also improves their properties, moving toward noise-robust architectures with better deployability. Using standard telecommunications infrastructure, we show an order of magnitude increase in the coincidence-to-accidental ratio for time-bin entangled photon pairs. Furthermore, we demonstrate the revival of lost entanglement by recovering quantum interference visibility and fidelity from quantum state tomography measurements of qubits corrupted by noise.
- 30 January 2026
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 12 (5) , eady8981
Open AccessQuantum state revival via coherent energy redistribution
Processing and detecting quantum states with high fidelity are essential for enabling quantum advantages across many applications. However, these states are known to be fragile because of their sensitivity to losses, their inability to be amplified, and their susceptibility to decoherence. This makes them far more vulnerable to noise than classical signals, limiting their out-of-lab deployment. We demonstrate quantum coherent energy redistribution, which not only recovers entangled states buried in noise but also improves their properties, moving toward noise-robust architectures with better deployability. Using standard telecommunications infrastructure, we show an order of magnitude increase in the coincidence-to-accidental ratio for time-bin entangled photon pairs. Furthermore, we demonstrate the revival of lost entanglement by recovering quantum interference visibility and fidelity from quantum state tomography measurements of qubits corrupted by noise.
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- p. 1-11
Talbot Array Illuminator Sonogram for Ultrafast Optical Waveforms
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- p. 1-11
Open AccessTalbot Array Illuminator Sonogram for Ultrafast Optical Waveforms
- Open Access
- 5 June 2025
- Published by American Physical Society (APS) in Physical Review Letters
- Vol. 134 (22) , 220801
Exploiting Nonlocal Correlations for Dispersion-Resilient Quantum Communications
Encoding quantum information via time-bin entangled states has had a profound impact on the development of quantum communications. However, dispersive propagation limits their achievable transmission distances. Here we describe a regime for nonlocal dispersion cancellation where the sum of arrival times of photons undergoing identical dispersion remains highly correlated. We exploit this effect to mitigate dispersive effects in a quantum key distribution fiber link, allowing an increase in the secret key rate by over a factor of 5 after 80 km of optical fiber dispersion. Published by the American Physical Society 2025
- 5 June 2025
- Published by American Physical Society (APS) in Physical Review Letters
- Vol. 134 (22) , 220801
Open AccessExploiting Nonlocal Correlations for Dispersion-Resilient Quantum Communications
Encoding quantum information via time-bin entangled states has had a profound impact on the development of quantum communications. However, dispersive propagation limits their achievable transmission distances. Here we describe a regime for nonlocal dispersion cancellation where the sum of arrival times of photons undergoing identical dispersion remains highly correlated. We exploit this effect to mitigate dispersive effects in a quantum key distribution fiber link, allowing an increase in the secret key rate by over a factor of 5 after 80 km of optical fiber dispersion. Published by the American Physical Society 2025
- 3 April 2025
- Published by Springer Nature in Nature Photonics
- Vol. 19 (4) , 340-343
Intertwining with Fourier optics
- 3 April 2025
- Published by Springer Nature in Nature Photonics
- Vol. 19 (4) , 340-343
Intertwining with Fourier optics
- Open Access
- 3 April 2025
- Published by Springer Nature in npj Nanophotonics
- Vol. 2 (1) , 1-21
Linear optical wave energy redistribution methods for photonic signal processing
Manipulating the phase of an optical wave over time and frequency gives full control to the user to implement a wide variety of energy preserving transformations directly in the analogue optical domain. These can be achieved using widely available linear mechanisms, such as temporal phase modulation and spectral phase filtering. The techniques based on these linear optical wave energy redistribution (OWER) methods are inherently energy efficient and have significant speed and bandwidth advantages over digital signal processing. We describe several recent OWER methods for optical signal processing, including denoising passive amplification, real-time spectrogram analysis, passive logic computing, and more. These functionalities are relevant whenever the signal is found on a classical or quantum optical wave, or could be upconverted from radio frequencies or microwaves, and they are of interest for a wide range of applications in telecommunications, sensing, metrology, biomedical imaging, and astronomy. The energy preservation of these methods makes them particularly interesting for quantum optics applications. Furthermore, many of the individual components have been demonstrated on-chip, enabling miniaturization for applications where size and weight are a main constraint.
- 3 April 2025
- Published by Springer Nature in npj Nanophotonics
- Vol. 2 (1) , 1-21
Open AccessLinear optical wave energy redistribution methods for photonic signal processing
Manipulating the phase of an optical wave over time and frequency gives full control to the user to implement a wide variety of energy preserving transformations directly in the analogue optical domain. These can be achieved using widely available linear mechanisms, such as temporal phase modulation and spectral phase filtering. The techniques based on these linear optical wave energy redistribution (OWER) methods are inherently energy efficient and have significant speed and bandwidth advantages over digital signal processing. We describe several recent OWER methods for optical signal processing, including denoising passive amplification, real-time spectrogram analysis, passive logic computing, and more. These functionalities are relevant whenever the signal is found on a classical or quantum optical wave, or could be upconverted from radio frequencies or microwaves, and they are of interest for a wide range of applications in telecommunications, sensing, metrology, biomedical imaging, and astronomy. The energy preservation of these methods makes them particularly interesting for quantum optics applications. Furthermore, many of the individual components have been demonstrated on-chip, enabling miniaturization for applications where size and weight are a main constraint.
- 3 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- Vol. 43 (14) , 6469-6483
Real-Time Spectrum Monitoring System for Next-Generation High-Capacity Optical Networks
This paper presents new theoretical and experimental results along with in-depth insight into a recently introduced simple real-time optical monitoring (RTOM) system and method, suitable for next-generation high-capacity optical networks. The proposed method employs electro-optic (EO) temporal sampling, followed by dispersion-induced real-time Fourier transformation and high-speed signal acquisition. The fundamentals, main design trade-offs, capabilities, and limitations of the RTOM system are thoroughly examined, including a detailed investigation of detection noise and sampling module extinction ratio (ER) effects on dynamic range. The RTOM design process is explored, and reasonable values for design parameters are identified based on practical and realistic specifications. The RTOM method continuously maps the spectral content of dense-wavelength-division-multiplexed (DWDM) data streams into the time domain, allowing to determine the presence and relative power of individual channels with high frequency resolution (∼30 GHz) and high sensitivity (∼1 dB). Notably, it provides fast measurement update rates (in the MHz range), far surpassing the measurement speed of commercially available optical spectrum analyzers, which is typically in the kHz range and slower. This scheme demonstrates remarkable versatility in performing real-time spectral analysis of DWDM signals across diverse operational conditions, including different modulation formats, bit rates (accommodating up to several hundred Gb/s per channel), various dispersion values in the optical link, and channel spacings from 50 GHz to 100 GHz spanning the entire C-band. The comprehensive evaluation reported here underscores the system's adaptability to diverse network configurations and transmission parameters, positioning it as a powerful tool for advanced optical network monitoring.
- 3 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- Vol. 43 (14) , 6469-6483
Real-Time Spectrum Monitoring System for Next-Generation High-Capacity Optical Networks
This paper presents new theoretical and experimental results along with in-depth insight into a recently introduced simple real-time optical monitoring (RTOM) system and method, suitable for next-generation high-capacity optical networks. The proposed method employs electro-optic (EO) temporal sampling, followed by dispersion-induced real-time Fourier transformation and high-speed signal acquisition. The fundamentals, main design trade-offs, capabilities, and limitations of the RTOM system are thoroughly examined, including a detailed investigation of detection noise and sampling module extinction ratio (ER) effects on dynamic range. The RTOM design process is explored, and reasonable values for design parameters are identified based on practical and realistic specifications. The RTOM method continuously maps the spectral content of dense-wavelength-division-multiplexed (DWDM) data streams into the time domain, allowing to determine the presence and relative power of individual channels with high frequency resolution (∼30 GHz) and high sensitivity (∼1 dB). Notably, it provides fast measurement update rates (in the MHz range), far surpassing the measurement speed of commercially available optical spectrum analyzers, which is typically in the kHz range and slower. This scheme demonstrates remarkable versatility in performing real-time spectral analysis of DWDM signals across diverse operational conditions, including different modulation formats, bit rates (accommodating up to several hundred Gb/s per channel), various dispersion values in the optical link, and channel spacings from 50 GHz to 100 GHz spanning the entire C-band. The comprehensive evaluation reported here underscores the system's adaptability to diverse network configurations and transmission parameters, positioning it as a powerful tool for advanced optical network monitoring.
- Open Access
- 6 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Photonics Journal
- Vol. 17 (2) , 1-8
Capacitively Coupled Silicon Modulator Fabricated on the Standard Silicon-on-Insulator Platform
Silicon-on-insulator (SOI) based silicon modulators are essential for modern optical communication links. Almost all these modulators are based on a rib waveguide structure. In this paper, we report our observation of a silicon modulator in a silicon strip waveguide. Such waveguide have a simpler structure, higher optical confinement, lower dispersion and can result in modulators with improved bandwidth compared to a rib waveguide. We demonstrated a novel strip waveguide-based Mach-Zehnder interferometer (MZI) silicon modulator fabricated in the Multi Project Wafer (MPW) run from a commercial foundry. Using a self-heterodyne measurement system we estimated a half-wave-voltage-length product (V $_{\pi }$ L) of $\sim$ 9.4 V.cm at 20 MHz modulation signal. The electro-optic (EO) measurement of the proposed modulator can generate the first order sidebands up to 68 GHz and almost a flat frequency response was observed.
- 6 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Photonics Journal
- Vol. 17 (2) , 1-8
Open AccessCapacitively Coupled Silicon Modulator Fabricated on the Standard Silicon-on-Insulator Platform
Silicon-on-insulator (SOI) based silicon modulators are essential for modern optical communication links. Almost all these modulators are based on a rib waveguide structure. In this paper, we report our observation of a silicon modulator in a silicon strip waveguide. Such waveguide have a simpler structure, higher optical confinement, lower dispersion and can result in modulators with improved bandwidth compared to a rib waveguide. We demonstrated a novel strip waveguide-based Mach-Zehnder interferometer (MZI) silicon modulator fabricated in the Multi Project Wafer (MPW) run from a commercial foundry. Using a self-heterodyne measurement system we estimated a half-wave-voltage-length product (V $_{\pi }$ L) of $\sim$ 9.4 V.cm at 20 MHz modulation signal. The electro-optic (EO) measurement of the proposed modulator can generate the first order sidebands up to 68 GHz and almost a flat frequency response was observed.
- Top Cited PapersOpen Access
- 2 January 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1-10
Quantum key distribution implemented with d-level time-bin entangled photons
High-dimensional photon states (qudits) are pivotal to enhance the information capacity, noise robustness, and data rates of quantum communications. Time-bin entangled qudits are promising candidates for implementing high-dimensional quantum communications over optical fiber networks with processing rates approaching those of classical telecommunications. However, their use is hindered by phase instability, timing inaccuracy, and low scalability of interferometric schemes needed for time-bin processing. As well, increasing the number of time bins per photon state typically requires decreasing the repetition rate of the system, affecting in turn the effective qudit rates. Here, we demonstrate a fiber-pigtailed, integrated photonic platform enabling the generation and processing of picosecond-spaced time-bin entangled qudits in the telecommunication C band via an on-chip interferometry system. We experimentally demonstrate the Bennett-Brassard-Mermin 1992 quantum key distribution protocol with time-bin entangled qudits and extend it over a 60 km-long optical fiber link, by showing dimensionality scaling without sacrificing the repetition rate. Our approach enables the manipulation of time-bin entangled qudits at processing speeds typical of standard telecommunications (10 s of GHz) with high quantum information capacity per single frequency channel, representing an important step towards an efficient implementation of high-data rate quantum communications in standard, multi-user optical fiber networks.
- 2 January 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1-10
Top Cited PapersOpen AccessQuantum key distribution implemented with d-level time-bin entangled photons
High-dimensional photon states (qudits) are pivotal to enhance the information capacity, noise robustness, and data rates of quantum communications. Time-bin entangled qudits are promising candidates for implementing high-dimensional quantum communications over optical fiber networks with processing rates approaching those of classical telecommunications. However, their use is hindered by phase instability, timing inaccuracy, and low scalability of interferometric schemes needed for time-bin processing. As well, increasing the number of time bins per photon state typically requires decreasing the repetition rate of the system, affecting in turn the effective qudit rates. Here, we demonstrate a fiber-pigtailed, integrated photonic platform enabling the generation and processing of picosecond-spaced time-bin entangled qudits in the telecommunication C band via an on-chip interferometry system. We experimentally demonstrate the Bennett-Brassard-Mermin 1992 quantum key distribution protocol with time-bin entangled qudits and extend it over a 60 km-long optical fiber link, by showing dimensionality scaling without sacrificing the repetition rate. Our approach enables the manipulation of time-bin entangled qudits at processing speeds typical of standard telecommunications (10 s of GHz) with high quantum information capacity per single frequency channel, representing an important step towards an efficient implementation of high-data rate quantum communications in standard, multi-user optical fiber networks.
- 21 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Microwave Theory and Techniques
- Vol. 73 (6) , 3424-3441
Versatile Photonic Spectrograms for Ultrafast Real-Time Broadband Microwave Signal Analysis
Joint time–frequency representations (JTFRs), such as the spectrogram, provide key information about the spectral evolution of a system. This is essential for instance to analyze telecommunication channels or sensing targets. Modern electronics cannot obtain JTFRs in real time without gaps for microwave signals with tens of gigahertz of instantaneous bandwidth and nanosecond temporal features. Previous photonics approaches are rigid in their permissible design specifications and cannot simultaneously address all performance requirements. We demonstrate the connection between two real-time, gapless, analog photonics spectrograms with versatile performance exceeding any previous technique, electronic, or otherwise. The time-lens spectrogram (TLS) and the Talbot array illuminator spectrogram (TAIS) are implemented via linear phase transformations using identical hardware to time-map the dynamic spectral information of microwave signals. The TLS enables analysis over broad bandwidths ( $>$ 100 GHz) with sharp time resolutions (tens of picoseconds), but a limited number of frequency analysis points. Conversely, the TAIS is limited in bandwidth and time resolution but allows for hundreds of frequency analysis points, enabling sub-gigahertz resolution. With these spectrograms, we measure high-speed transients with sub-nanosecond time resolution, multichannel nanosecond frequency hopping communication signals, and signals with bandwidths much higher than that of the detector and analog-to-digital converter employed.
- 21 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Microwave Theory and Techniques
- Vol. 73 (6) , 3424-3441
Versatile Photonic Spectrograms for Ultrafast Real-Time Broadband Microwave Signal Analysis
Joint time–frequency representations (JTFRs), such as the spectrogram, provide key information about the spectral evolution of a system. This is essential for instance to analyze telecommunication channels or sensing targets. Modern electronics cannot obtain JTFRs in real time without gaps for microwave signals with tens of gigahertz of instantaneous bandwidth and nanosecond temporal features. Previous photonics approaches are rigid in their permissible design specifications and cannot simultaneously address all performance requirements. We demonstrate the connection between two real-time, gapless, analog photonics spectrograms with versatile performance exceeding any previous technique, electronic, or otherwise. The time-lens spectrogram (TLS) and the Talbot array illuminator spectrogram (TAIS) are implemented via linear phase transformations using identical hardware to time-map the dynamic spectral information of microwave signals. The TLS enables analysis over broad bandwidths ( $>$ 100 GHz) with sharp time resolutions (tens of picoseconds), but a limited number of frequency analysis points. Conversely, the TAIS is limited in bandwidth and time resolution but allows for hundreds of frequency analysis points, enabling sub-gigahertz resolution. With these spectrograms, we measure high-speed transients with sub-nanosecond time resolution, multichannel nanosecond frequency hopping communication signals, and signals with bandwidths much higher than that of the detector and analog-to-digital converter employed.