-
Sub-20 kHz low frequency noise near ultraviolet miniature external cavity laser diode
Authors:
R Kervazo,
A Congar,
G Perin,
L Lablonde,
R Butté,
N Grandjean,
L Bodiou,
J Charrier,
S Trebaol
Abstract:
We present a compact InGaN fiber Bragg grating (FBG) semiconductor laser diode operating below 400 nm in the single-mode emission regime. This compact coherent laser source exhibits an intrinsic linewidth of 14 kHz in the near-UV range and a side-mode suppression ratio reaching up to 40 dB accompanied by a mW-level output power. Furthermore, the properties of the FBG, including its central wavelen…
▽ More
We present a compact InGaN fiber Bragg grating (FBG) semiconductor laser diode operating below 400 nm in the single-mode emission regime. This compact coherent laser source exhibits an intrinsic linewidth of 14 kHz in the near-UV range and a side-mode suppression ratio reaching up to 40 dB accompanied by a mW-level output power. Furthermore, the properties of the FBG, including its central wavelength, bandwidth, and reflectivity, can be readily customized to fulfill specific requirements. As a result, the small footprint design of this laser is compatible with integration into a standard butterfly package to ease the lab-to-market technology transfer. The combination of low frequency noise and fibered output signal positions these FBG laser systems as strong candidates for hybridization with integrated photonic platforms tailored for quantum information processing and metrology.
△ Less
Submitted 24 May, 2024;
originally announced May 2024.
-
VREM-FL: Mobility-Aware Computation-Scheduling Co-Design for Vehicular Federated Learning
Authors:
Luca Ballotta,
Nicolò Dal Fabbro,
Giovanni Perin,
Luca Schenato,
Michele Rossi,
Giuseppe Piro
Abstract:
Assisted and autonomous driving are rapidly gaining momentum and will soon become a reality. Artificial intelligence and machine learning are regarded as key enablers thanks to the massive amount of data that smart vehicles will collect from onboard sensors. Federated learning is one of the most promising techniques for training global machine learning models while preserving data privacy of vehic…
▽ More
Assisted and autonomous driving are rapidly gaining momentum and will soon become a reality. Artificial intelligence and machine learning are regarded as key enablers thanks to the massive amount of data that smart vehicles will collect from onboard sensors. Federated learning is one of the most promising techniques for training global machine learning models while preserving data privacy of vehicles and optimizing communications resource usage. In this article, we propose vehicular radio environment map federated learning (VREM-FL), a computation-scheduling co-design for vehicular federated learning that combines mobility of vehicles with 5G radio environment maps. VREM-FL jointly optimizes learning performance of the global model and wisely allocates communication and computation resources. This is achieved by orchestrating local computations at the vehicles in conjunction with transmission of their local models in an adaptive and predictive fashion, by exploiting radio channel maps. The proposed algorithm can be tuned to trade training time for radio resource usage. Experimental results demonstrate that VREM-FL outperforms literature benchmarks for both a linear regression model (learning time reduced by 28%) and a deep neural network for semantic image segmentation (doubling the number of model updates within the same time window).
△ Less
Submitted 3 June, 2024; v1 submitted 30 November, 2023;
originally announced November 2023.
-
Static and Dynamic Jamming Games Over Wireless Channels With Mobile Strategic Players
Authors:
Giovanni Perin,
Leonardo Badia
Abstract:
We study a wireless jamming problem consisting of the competition between a legitimate receiver and a jammer, as a zero-sum game with the value to maximize/minimize being the channel capacity at the receiver's side. Most of the approaches found in the literature consider the two players to be stationary nodes. Instead, we investigate what happens when they can change location, specifically moving…
▽ More
We study a wireless jamming problem consisting of the competition between a legitimate receiver and a jammer, as a zero-sum game with the value to maximize/minimize being the channel capacity at the receiver's side. Most of the approaches found in the literature consider the two players to be stationary nodes. Instead, we investigate what happens when they can change location, specifically moving along a linear geometry. We frame this at first as a static game, which can be solved in closed form, and subsequently we extend it to a dynamic game, under three different versions for what concerns completeness/perfection of mutual information about the adversary's position, corresponding to different assumptions of concealment/sequentiality of the moves, respectively. We first provide some theoretical conditions that hold for the static game and also help identify good strategies valid under any setup, including dynamic games. Since dynamic games, although more realistic, are characterized by an exploding strategy space, we exploit reinforcement learning to obtain efficient strategies leading to equilibrium outcomes. We show how theoretical findings can be used to train smart agents to play the game, and validate our approach in practical setups.
△ Less
Submitted 19 June, 2023;
originally announced June 2023.
-
IB-RAR: Information Bottleneck as Regularizer for Adversarial Robustness
Authors:
Xiaoyun Xu,
Guilherme Perin,
Stjepan Picek
Abstract:
In this paper, we propose a novel method, IB-RAR, which uses Information Bottleneck (IB) to strengthen adversarial robustness for both adversarial training and non-adversarial-trained methods. We first use the IB theory to build regularizers as learning objectives in the loss function. Then, we filter out unnecessary features of intermediate representation according to their mutual information (MI…
▽ More
In this paper, we propose a novel method, IB-RAR, which uses Information Bottleneck (IB) to strengthen adversarial robustness for both adversarial training and non-adversarial-trained methods. We first use the IB theory to build regularizers as learning objectives in the loss function. Then, we filter out unnecessary features of intermediate representation according to their mutual information (MI) with labels, as the network trained with IB provides easily distinguishable MI for its features. Experimental results show that our method can be naturally combined with adversarial training and provides consistently better accuracy on new adversarial examples. Our method improves the accuracy by an average of 3.07% against five adversarial attacks for the VGG16 network, trained with three adversarial training benchmarks and the CIFAR-10 dataset. In addition, our method also provides good robustness for undefended methods, such as training with cross-entropy loss only. Finally, in the absence of adversarial training, the VGG16 network trained using our method and the CIFAR-10 dataset reaches an accuracy of 35.86% against PGD examples, while using all layers reaches 25.61% accuracy.
△ Less
Submitted 31 May, 2023; v1 submitted 9 February, 2023;
originally announced February 2023.
-
Divide and Save: Splitting Workload Among Containers in an Edge Device to Save Energy and Time
Authors:
Aria Khoshsirat,
Giovanni Perin,
Michele Rossi
Abstract:
The increasing demand for edge computing is leading to a rise in energy consumption from edge devices, which can have significant environmental and financial implications. To address this, in this paper we present a novel method to enhance the energy efficiency while speeding up computations by distributing the workload among multiple containers in an edge device. Experiments are conducted on two…
▽ More
The increasing demand for edge computing is leading to a rise in energy consumption from edge devices, which can have significant environmental and financial implications. To address this, in this paper we present a novel method to enhance the energy efficiency while speeding up computations by distributing the workload among multiple containers in an edge device. Experiments are conducted on two Nvidia Jetson edge boards, the TX2 and the AGX Orin, exploring how using a different number of containers can affect the energy consumption and the computational time for an inference task. To demonstrate the effectiveness of our splitting approach, a video object detection task is conducted using an embedded version of the state-of-the-art YOLO algorithm, quantifying the energy and the time savings achieved compared to doing the computations on a single container. The proposed method can help mitigate the environmental and economic consequences of high energy consumption in edge computing, by providing a more sustainable approach to managing the workload of edge devices.
△ Less
Submitted 23 March, 2023; v1 submitted 13 February, 2023;
originally announced February 2023.
-
High-Q Whispering-Gallery-Modes Microresonators in the Near-Ultraviolet Spectral Range
Authors:
Georges Perin,
Yannick Dumeige,
Patrice Féron,
Stéphane Trebaol
Abstract:
High-$Q$ whispering gallery mode microresonators are cornerstones to develop compact coherent sources like single frequency lasers or frequency combs. This work reports a detailed characterization of a whispering gallery mode microsphere in the near ultraviolet. Light coupling in the resonator is obtain thanks to a robust angle-polished fiber allowing to investigate the different coupling regimes.…
▽ More
High-$Q$ whispering gallery mode microresonators are cornerstones to develop compact coherent sources like single frequency lasers or frequency combs. This work reports a detailed characterization of a whispering gallery mode microsphere in the near ultraviolet. Light coupling in the resonator is obtain thanks to a robust angle-polished fiber allowing to investigate the different coupling regimes. An intrinsic $Q_0$-factor of $2.2 \times 10^8$ and a finesse of $7.3 \times 10^4$ are reported at a wavelength of 420 nm. Physical mechanisms contributing to the $Q_0$-factor are discussed and routes to improve the performances are drawn. Such high-$Q$ factor and high finesse are key ingredients to ease the study of photonic devices based on WGM microresonators.
△ Less
Submitted 7 November, 2022;
originally announced November 2022.
-
Energy Efficient Deployment and Orchestration of Computing Resources at the Network Edge: a Survey on Algorithms, Trends and Open Challenges
Authors:
Neda Shalavi,
Giovanni Perin,
Andrea Zanella,
Michele Rossi
Abstract:
Mobile networks are becoming energy hungry, and this trend is expected to continue due to a surge in communication and computation demand. Multi-access Edge Computing (MEC), will entail energy-consuming services and applications, with non-negligible impact in terms of ecological sustainability. In this paper, we provide a comprehensive review of existing approaches to make edge computing networks…
▽ More
Mobile networks are becoming energy hungry, and this trend is expected to continue due to a surge in communication and computation demand. Multi-access Edge Computing (MEC), will entail energy-consuming services and applications, with non-negligible impact in terms of ecological sustainability. In this paper, we provide a comprehensive review of existing approaches to make edge computing networks greener, including but not limited to the exploitation of renewable energy resources, and context-awareness. We hence provide an updated account of recent developments on MEC from an energetic sustainability perspective, addressing the initial deployment of computing resources, their dynamic (re)allocation, as well as distributed and federated learning designs. In doing so, we highlight the energy aspects of these algorithms, advocating the need for energy-sustainable edge computing systems that are aligned with Sustainable Development Goals (SDGs) and the Paris agreement. To the best of our knowledge, this is the first work providing a systematic literature review on the efficient deployment and management of energy harvesting MEC, with special focus on the deployment, provisioning, and scheduling of computing tasks, including federated learning for distributed edge intelligence, toward making edge networks greener and more sustainable. At the end of the paper, open research avenues and challenges are identified for all the surveyed topics.
△ Less
Submitted 28 September, 2022;
originally announced September 2022.
-
EASE: Energy-Aware Job Scheduling for Vehicular Edge Networks With Renewable Energy Resources
Authors:
Giovanni Perin,
Francesca Meneghello,
Ruggero Carli,
Luca Schenato,
Michele Rossi
Abstract:
The energy sustainability of multi-access edge computing (MEC) platforms is here addressed by develo** Energy-Aware job Scheduling at the Edge (EASE), a computing resource scheduler for edge servers co-powered by renewable energy resources and the power grid. The scenario under study involves the optimal allocation and migration of time-sensitive computing tasks in a resource-constrained interne…
▽ More
The energy sustainability of multi-access edge computing (MEC) platforms is here addressed by develo** Energy-Aware job Scheduling at the Edge (EASE), a computing resource scheduler for edge servers co-powered by renewable energy resources and the power grid. The scenario under study involves the optimal allocation and migration of time-sensitive computing tasks in a resource-constrained internet of vehicles (IoV) context. This is achieved by tackling, as the main objective, the minimization of the carbon footprint of the edge network, whilst delivering adequate quality of service (QoS) to the end users (e.g., meeting task execution deadlines). EASE integrates i) a centralized optimization step, solved through model predictive control (MPC), to manage the renewable energy that is locally collected at the edge servers and their local computing resources, estimating their future availability, and ii) a distributed consensus step, solved via dual ascent in closed form, to reach agreement on service migrations. EASE is compared with four existing migration strategies. Quantitative results demonstrate its greater energy efficiency, which often gets close to complete carbon neutrality, while also improving the QoS.
△ Less
Submitted 12 August, 2022; v1 submitted 3 November, 2021;
originally announced November 2021.
-
Narrow Linewidth near-UV InGaN Laser Diode based on External Cavity Fiber Bragg Grating
Authors:
Antoine Congar,
Mathilde Gay,
Georges Perin,
Dominique Mammez,
Jean-Claude Simon,
Pascal Besnard,
Julien Rouvillain,
Thierry Georges,
Laurent Lablonde,
Thierry Robin,
Stéphane Trebaol
Abstract:
We realize a fiber Bragg grating InGaN based laser diode emitting at 400 nm and demonstrate its high coherency. Thanks to the fabrication of a narrow band fiber Bragg grating in the near-UV, we can reach single-mode and single-frequency regimes for the self-injection locked diode. The device exhibits 44 dB side-mode-suppression-ratio and mW output power. Detailed frequency noise analysis reveals s…
▽ More
We realize a fiber Bragg grating InGaN based laser diode emitting at 400 nm and demonstrate its high coherency. Thanks to the fabrication of a narrow band fiber Bragg grating in the near-UV, we can reach single-mode and single-frequency regimes for the self-injection locked diode. The device exhibits 44 dB side-mode-suppression-ratio and mW output power. Detailed frequency noise analysis reveals sub-MHz integrated linewidth and 16 kHz intrinsic linewidth. Such a narrow linewidth laser diode in the near-UV domain with a compact and low-cost design could find applications whenever coherency and interferometric resolutions are needed.
△ Less
Submitted 8 February, 2021;
originally announced February 2021.
-
Impact of cation redox chemistry on continuous hydrothermal synthesis of 2D-Ni(Co/Fe) hydroxides
Authors:
Massimo Rosa,
Debora Marani,
Giovanni Perin,
Søren Bredmose Simonsen,
Philipp Zielke,
Antonella Glisenti,
Ragnar Kiebach,
Andreas Lesch,
Vincenzo Esposito
Abstract:
Continuous hydrothermal flow synthesis (CHFS) is a facile, upscalable and cost-efficient synthetic method enabling the nanostructuring of advanced functional materials in steady conditions, i.e. not in batch synthesis. In this paper, we use CHFS to crystallize NiCo- and NiFe-hydroxides in water solution with 2D nanofeatures. By tuning the synthetic parameters, we disclose the key role of the catio…
▽ More
Continuous hydrothermal flow synthesis (CHFS) is a facile, upscalable and cost-efficient synthetic method enabling the nanostructuring of advanced functional materials in steady conditions, i.e. not in batch synthesis. In this paper, we use CHFS to crystallize NiCo- and NiFe-hydroxides in water solution with 2D nanofeatures. By tuning the synthetic parameters, we disclose the key role of the cation redox chemistry in the transition between two competitive phases: from 2D-nanoplatelets of brucite to layered double hydroxides (LDH). For controlling the precipitation of different Ni, Fe, Co-hydroxide phases, we propose the combined use of an oxidizing (H2O2) and a complexing (NH3) agent. At temperatures as low as 80 °C, the presence of H2O2 and a low concentration of NH3 favour the Ni2+/Co3+ over Ni2+/Co2+ oxidation states, shifting the product structure from brucite phase (temperatures > 80 °C) to LDH. Conversely, for the NiFe-hydroxides the transition from LDH (temperatures < 80 °C) to brucite phase (temperatures > 80 °C) is controlled by the reaction temperature only. Due to the high stability of Fe3+, the synthesis of NiFe products by CHFS does not require oxidizing and complexing agents, resulting in a robust process for large-scale production.
△ Less
Submitted 11 October, 2019;
originally announced October 2019.
-
Gd0.2Ce0.8O1.9/Y0.16Zr0.84O1.92 nanocomposite thin films for low temperature ionic conductivity
Authors:
Giovanni Perin,
Christophe Gadea,
Massimo Rosa,
Simone Sanna,
Yu Xu,
Ragnar Kiebach,
Antonella Glisenti,
Vincenzo Esposito
Abstract:
Gd0.2Ce0.8O1.9/Y0.16Zr0.84O1.92 (GDC/YSZ) nanocomposite is synthesized by a novel hybrid chemical route, where colloidal crystalline GDC nanoparticles from continuous hydrothermal synthesis are dispersed into a metalorganic YSZ matrix precursor. The result is a mixture of metal oxides in which GDC nanoparticles are finely distributed in a continuous metalorganic polymeric matrix to be crystallized…
▽ More
Gd0.2Ce0.8O1.9/Y0.16Zr0.84O1.92 (GDC/YSZ) nanocomposite is synthesized by a novel hybrid chemical route, where colloidal crystalline GDC nanoparticles from continuous hydrothermal synthesis are dispersed into a metalorganic YSZ matrix precursor. The result is a mixture of metal oxides in which GDC nanoparticles are finely distributed in a continuous metalorganic polymeric matrix to be crystallized after calcination. The GDC nanoparticles reduce the temperature necessary to obtain crystalline YSZ, which is already formed at 400 °C. The nanocomposite reveals structural stability up to 800 °C when treated in both air and reducing atmosphere, showing the onset of diffusion below 1000 °C. The diffusional processes are largely dependent on the nanometric grain size, with Zr4+ diffusing abruptly towards GDC in air at 1000 °C and GDC/YSZ interdiffusion being hindered in reducing environment despite the onset temperature of 900 °C. The nanocomposite precursor is an inkjet-printable reactive water-based material, suitable for the deposition of thin films with a thickness below 100 nm after calcination at 750 °C. The crystal structure of the film reveals no interaction between GDC and YSZ but a microstrain (0.3% tensile strain for YSZ). The thin film microstructure shows a compact layer with 94% density. The nanocomposite shows high oxygen ionic conductivity at low temperatures (> 5 * 10-3 S cm-1 at 500 °C), low activation energy (0.55 eV), and dominant oxygen ionic conductivity even in reducing conditions (pO2 < 10-25 atm). We show that these properties arise from the large interface between the components of the composite, due to the embedding of the GDC nanoparticles in the YSZ matrix, while ZrO-CeO intermixing can be avoided and no n-type conductivity is observed even at low oxygen activities and high temperatures.
△ Less
Submitted 15 April, 2019;
originally announced April 2019.
