Related references
Note: Only part of the references are listed.
Article
Engineering, Electrical & Electronic
Uiara C. de Moura et al.
Summary: In this paper, a fiber-agnostic neural network model is proposed to predict the gain profile of a new fiber type based on experimental data from different optical fibers. The results show that a wide range of optical fiber parameters is necessary to accurately predict the gain of a new fiber. Additionally, the use of transfer learning is proposed to improve the general model's accuracy with limited experimental measurements.
JOURNAL OF LIGHTWAVE TECHNOLOGY
(2023)
Article
Engineering, Electrical & Electronic
Ali Cem et al.
Summary: Photonic integrated circuits are advancing optical neural networks, which could potentially outperform electronic counterparts in speed and energy efficiency due to the suitability of optical signals for matrix multiplication. However, programming photonic chips for accurate optical matrix multiplication remains challenging.
JOURNAL OF LIGHTWAVE TECHNOLOGY
(2023)
Article
Engineering, Electrical & Electronic
Irene Estebanez et al.
Summary: Analog photonic computing has been proposed and tested as an alternative approach for data recovery in fiber transmission systems. This study demonstrates the effectiveness of internal fading memory in photonic reservoir computing, highlighting its dependence on signal properties. It also compares data recovery performances between reservoir computing and extreme learning machine fiber-based configurations, showing equivalent results and the advantage of simplified system and increased computation speed with the latter method.
JOURNAL OF LIGHTWAVE TECHNOLOGY
(2022)
Review
Physics, Multidisciplinary
Chaoran Huang et al.
Summary: Neuromorphic engineering using photonics allows for high-speed and energy-efficient artificial intelligence and neuromorphic computing applications. It mimics neurons and synapses in the brain to achieve distributed and parallel processing.
ADVANCES IN PHYSICS-X
(2022)
Article
Optics
Bin Shi et al.
Summary: This paper demonstrates the use of a wavelength converter based on cross-gain modulation in a semiconductor optical amplifier (SOA) as a nonlinear function integrated within an all-optical neuron. The impact of fully integrated linear and nonlinear functions on the neuron output is investigated. The performance of the monolithically integrated neuron is found to be better than that of the corresponding hybrid device in terms of accuracy. These all-optical neurons are then used to simulate a two-layer photonic deep neural network for handwritten digit classification, achieving high accuracy and energy efficiency compared to state-of-the-art graphics processing units. The importance of scaling photonic integrated neural networks on chip is emphasized.
Review
Optics
Wei Ma et al.
Summary: Innovative approaches and tools, particularly deep learning, are shaping the field of photonics by offering efficient means to design photonic structures and providing data-driven solutions complementary to traditional physics-based methods. The progress in deep-learning-based photonic design is promising, with various model architectures showing potential applications in specific photonic tasks.
Article
Engineering, Electrical & Electronic
Stenio M. Ranzini et al.
Summary: The research offers a solution for low latency and reliable connections in the network of cloud edge data centers, overcoming the main limitation of chromatic dispersion for high symbol rate systems. By utilizing a receiver with shared complexity in optical and digital domains, the goal of achieving 80 km transmission at 32 GBd signal has been successfully demonstrated.
JOURNAL OF LIGHTWAVE TECHNOLOGY
(2021)
Article
Chemistry, Multidisciplinary
Lorenzo De Marinis et al.
Summary: This paper characterizes and compares two thermally tuned photonic integrated processors, based on silicon-on-insulator and silicon nitride platforms, for feature map extraction in convolutional neural networks. The reduction in bit resolution when crossing the processor is mainly due to optical losses, with the silicon-on-insulator chip ranging from 2.3-3.3 and the silicon nitride chip ranging from 1.3-2.4. However, the lower extinction ratio of Mach-Zehnder elements in the latter platform limits their expressivity to 75% compared to the former. Ultimately, the silicon-on-insulator processor outperforms the silicon nitride one in terms of footprint and energy efficiency.
APPLIED SCIENCES-BASEL
(2021)
Proceedings Paper
Optics
F. Da Ros et al.
Summary: The substantial increase in communication throughput driven by machine-to-machine communication within and between data centers is straining short-reach communication links. Researchers are investigating equalization techniques for intensity-modulation/direct-detection transmission, focusing on addressing issues like inter-symbol interference introduced by fiber dispersion. Complex approaches involving maximum likelihood sequence estimation and machine-learning tools, such as neural networks, are being explored to improve performance.
METRO AND DATA CENTER OPTICAL NETWORKS AND SHORT-REACH LINKS IV
(2021)
Article
Engineering, Electrical & Electronic
Darko Zibar et al.
JOURNAL OF LIGHTWAVE TECHNOLOGY
(2020)
Article
Engineering, Electrical & Electronic
Francesco Da Ros et al.
IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS
(2020)
Article
Engineering, Electrical & Electronic
Maziyar Milanizadeh et al.
IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS
(2020)
Article
Multidisciplinary Sciences
Daniel Perez-Lopez et al.
NATURE COMMUNICATIONS
(2020)
Article
Engineering, Electrical & Electronic
Andrew Katumba et al.
JOURNAL OF LIGHTWAVE TECHNOLOGY
(2019)
Article
Optics
Boris Karanov et al.
News Item
Optics
Darko Zibar et al.
Article
Multidisciplinary Sciences
Daniel Perez et al.
NATURE COMMUNICATIONS
(2017)