no code implementations • 19 Mar 2024 • Isidora Teofilovic, Ali Cem, David Sanchez-Jacome, Daniel Perez-Lopez, Francesco Da Ros
Here, we train and experimentally evaluate three models incorporating varying degrees of physics intuition to predict the effect of thermal crosstalk in different locations of an integrated programmable photonic mesh.
no code implementations • 7 Dec 2023 • Bernard J. Giron Castro, Christophe Peucheret, Francesco Da Ros
We numerically demonstrate a silicon add-drop microring-based reservoir computing scheme that combines parallel delayed inputs and wavelength division multiplexing.
no code implementations • 10 Nov 2023 • Metodi Plamenov Yankov, Smaranika Swain, Ognjen Jovanovic, Darko Zibar, Francesco Da Ros
The GCS scheme is experimentally demonstrated in a multi-span recirculating loop coherent optical fiber transmission system with a total distance of up to 3000 km.
no code implementations • 25 Oct 2023 • Bernard J. Giron Castro, Christophe Peucheret, Darko Zibar, Francesco Da Ros
We numerically demonstrate a microring-based time-delay reservoir computing scheme that simultaneously solves three tasks involving time-series prediction, classification, and wireless channel equalization.
no code implementations • 13 Oct 2023 • Bernard J. Giron Castro, Christophe Peucheret, Darko Zibar, Francesco Da Ros
Microring resonators (MRRs) are promising devices for time-delay photonic reservoir computing, but the impact of the different physical effects taking place in the MRRs on the reservoir computing performance is yet to be fully understood.
no code implementations • 27 Sep 2023 • Sergio Hernandez, Ognjen Jovanovic, Christophe Peucheret, Francesco Da Ros, Darko Zibar
End-to-end learning has become a popular method for joint transmitter and receiver optimization in optical communication systems.
no code implementations • 11 Sep 2023 • Metodi P. Yankov, Mehran Soltani, Andrea Carena, Darko Zibar, Francesco Da Ros
Designing and optimizing optical amplifiers to maximize system performance is becoming increasingly important as optical communication systems strive to increase throughput.
no code implementations • 28 Aug 2023 • Yevhenii Osadchuk, Ognjen Jovanovic, Stenio M. Ranzini, Roman Dischler, Vahid Aref, Darko Zibar, Francesco Da Ros
In this work, we propose a low-complexity NN that performs samples-to-symbol equalization, meaning that the NN-based equalizer includes match filtering and downsampling.
no code implementations • 10 Aug 2023 • Ali Cem, Ognjen Jovanovic, Siqi Yan, Yunhong Ding, Darko Zibar, Francesco Da Ros
We present and experimentally evaluate using transfer learning to address experimental data scarcity when training neural network (NN) models for Mach-Zehnder interferometer mesh-based optical matrix multipliers.
no code implementations • 19 Jul 2023 • Metodi P. Yankov, Ognjen Jovanovic, Darko Zibar, Francesco Da Ros
A many-to-one mapping geometric constellation shaping scheme is proposed with a fixed modulation format, fixed FEC engine and rate adaptation with an arbitrarily small step.
no code implementations • 13 Jul 2023 • Bernard J. Giron Castro, Christophe Peucheret, Darko Zibar, Francesco Da Ros
We quantify the impact of thermo-optic and free-carrier effects on time-delay reservoir computing using a silicon microring resonator.
no code implementations • 15 May 2023 • Sergio Hernandez Fernandez, Christophe Peucheret, Ognjen Jovanovic, Francesco Da Ros, Darko Zibar
The end-to-end optimization of links based on directly-modulated lasers may require an analytically differentiable channel.
no code implementations • 29 Nov 2022 • Yevhenii Osadchuk, Ognjen Jovanovic, Darko Zibar, Francesco Da Ros
We propose spectrum-sliced reservoir computer-based (RC) multi-symbol equalization for 32-GBd PAM4 transmission.
no code implementations • 29 Nov 2022 • Ali Cem, Ognjen Jovanovic, Siqi Yan, Yunhong Ding, Darko Zibar, Francesco Da Ros
We demonstrate transfer learning-assisted neural network models for optical matrix multipliers with scarce measurement data.
no code implementations • 17 Oct 2022 • Ali Cem, Siqi Yan, Yunhong Ding, Darko Zibar, Francesco Da Ros
Photonic integrated circuits are facilitating the development of optical neural networks, which have the potential to be both faster and more energy efficient than their electronic counterparts since optical signals are especially well-suited for implementing matrix multiplications.
no code implementations • 26 Sep 2022 • Mehran Soltani, Francesco Da Ros, Andrea Carena, Darko Zibar
In this case, the experimental results assert that for 2D profiles with the target flat gain levels, the DE obtains less than 1 dB maximum gain deviation, when the setup is not physically limited in the pump power values.
no code implementations • 13 Jun 2022 • Metodi Plamenov Yankov, Francesco Da Ros, Uiara Celine de Moura, Andrea Carena, Darko Zibar
The forward propagation model is combined with an experimentally-trained ML model of a backward-pumping Raman amplifier to jointly optimize the frequency and power of the forward amplifier's pumps and the powers of the backward amplifier's pumps.
no code implementations • 16 May 2022 • Mehran Soltani, Francesco Da Ros, Andrea Carena, Darko Zibar
We experimentally validate a machine learning-enabled Raman amplification framework, capable of jointly shaping the signal power evolution in two domains: frequency and fiber distance.
no code implementations • 27 Jan 2022 • Metodi P. Yankov, Francesco Da Ros, Søren Forchhammer, Lars Gruner-Nielsen
The optical fiber multiple-input multiple-output (MIMO) channel with intensity modulation and direct detection (IM/DD) per spatial path is treated.
no code implementations • 23 Nov 2021 • Ali Cem, Siqi Yan, Uiara Celine de Moura, Yunhong Ding, Darko Zibar, Francesco Da Ros
We experimentally compare simple physics-based vs. data-driven neural-network-based models for offline training of programmable photonic chips using Mach-Zehnder interferometer meshes.
no code implementations • 16 Nov 2021 • Ognjen Jovanovic, Metodi P. Yankov, Francesco Da Ros, Darko Zibar
Two noise models are considered for the additive noise: white Gaussian noise and nonlinear interference noise model for fiber nonlinearities.
no code implementations • 7 Jun 2021 • Metodi Plamenov Yankov, Pawel Marcin Kaminski, Henrik Enggaard Hansen, Francesco Da Ros
When the input power profile is optimized for flat and maximized received SNR per channel, the minimum performance in an arbitrary 3-span experimental system is improved by up to 8 dB w. r. t.
no code implementations • 1 Jun 2021 • Ognjen Jovanovic, Metodi P. Yankov, Francesco Da Ros, Darko Zibar
We propose an autoencoder-based geometric shaping that learns a constellation robust to SNR and laser linewidth estimation errors.
no code implementations • 21 Feb 2021 • Mehran Soltani, Francesco Da Ros, Andrea Carena, Darko Zibar
We present a Convolutional Neural Network (CNN) architecture for inverse Raman amplifier design.
no code implementations • 21 Dec 2020 • Ognjen Jovanovic, Metodi Plamenov Yankov, Francesco Da Ros, Darko Zibar
Our results indicate that the autoencoder can be successfully optimized using the proposed training method to achieve better robustness to residual phase noise with respect to standard constellation schemes such as Quadrature Amplitude Modulation and Iterative Polar Modulation for the considered conditions.
no code implementations • 9 Dec 2020 • Uiara Celine de Moura, Ann Margareth Rosa Brusin, Andrea Carena, Darko Zibar, Francesco Da Ros
A machine learning framework predicting pump powers and noise figure profile for a target distributed Raman amplifier gain profile is experimentally demonstrated.
Applied Physics Optics
no code implementations • 8 Oct 2020 • Stenio M. Ranzini, Roman Dischler, Francesco Da Ros, Henning Buelow, Darko Zibar
A receiver with shared complexity between optical and digital domains is experimentally demonstrated.
no code implementations • 11 Sep 2020 • Simone Gaiarin, Francesco Da Ros, Rasmus T. Jones, Darko Zibar
In this first numerical analysis, the detection is performed by a neural network (NN), whereas the symbol-to-waveform mapping is aided by the nonlinear Fourier transform (NFT) theory in order to simplify and guide the optimization on the modulation side.
1 code implementation • 11 Sep 2020 • Francesco Da Ros, Uiara Celine de Moura, Metodi P. Yankov
We report a neural-network based erbium-doped fiber amplifier (EDFA) gain model built from experimental measurements.
1 code implementation • 11 Sep 2020 • Metodi P. Yankov, Uiara Celine de Moura, Francesco Da Ros
Cascades of a machine learning-based EDFA gain model trained on a single physical device and a fully differentiable stimulated Raman scattering fiber model are used to predict and optimize the power profile at the output of an experimental multi-span fully-loaded C-band optical communication system.