HAL-CentraleSupelec
Not a member yet
77624 research outputs found
Sort by
Estimation des DOA avec un réseau d’antennes coprime en présence d’erreurs de gain et de phase des capteurs
No full textNational audienceLes réseaux linéaires coprimes (CLAs) permettentd’estimer un nombre de sources supérieur aunombre de capteurs, contrairement aux réseaux linéairesuniformes (ULAs). Cependant, les méthodes de sousespaceet les approches parcimonieuses fonctionnent ensupposant une connaissance exacte de la géométrie du réseau,notamment des gains et phases des capteurs. En pratique,ces paramètres sont souvent perturbés par rapportà leurs valeurs nominales, ce qui dégrade fortement lesperformances des algorithmes d’estimation des directionsd’arrivées (DOAs). Dans cet article, nous proposons uneapproche basée sur Sparse Bayesian Learning (SBL) permettantd’estimer à la fois les DOA et les erreurs de gainet de phase du réseau. Nous introduisons également unenouvelle formule de mise à jour des perturbations afin deréduire la complexité calculatoire, ainsi qu’un mécanismeCFAR (Constant false-alarm rate) améliorant la précisionde l’estimation des DOA
Who decides? Positioning stakeholders in a decision-making process via network modeling and analysis: Application to regional trains decarbonization
No full textInternational audienceUnderstanding and controlling stakeholder influence in public–private decision-making remains a significant challenge for public authorities. This study proposes a methodology for diagnosing, modeling, and reorganizing stakeholder contributions within complex decision-making processes. The methodology integrates five steps: stakeholder identification, stakeholder network modeling, network analysis, stakeholder clustering, and a reorganization phase that aligns stakeholder involvement with the specific objectives of each decision stage. The methodology is applied to public-private network data to reveal influence patterns, dominant actors, and structural coalitions. Stakeholders are identified, and their interactions are formalized using a weighted Design Structure Matrix (DSM). Network indicators are computed to characterize interaction intensity and connectivity. Clustering algorithms are applied to identify clusters of closely interacting stakeholders, which serve as the basis for reorganizing stakeholder involvement across decision stages. We applied the methodology to the decarbonization of French regional diesel trains. Data were collected through interviews conducted with multiple regional transport authorities. The results show an intense concentration of decisional weight among industrial stakeholders and higher-level public institutions, indicating governance vulnerabilities stemming from asymmetric access to information and influence. A validation phase was conducted with a regional decision-maker (Corse), confirming the practical relevance of the proposed stakeholder reorganization and supporting its applicability to real-world decision processes. The study demonstrates that dynamically structuring stakeholder involvement can mitigate risks, enhance decision robustness, and assist public authorities in managing private-sector influence. While illustrated through a specific case study, the methodology is generalizable to any multi-actor public–private decision-making context
A Complex-Valued Continuous-Variable Quantum Approximation Optimization Algorithm (CCV-QAOA)
No full textContinuous-variable (CV) quantum systems offer a natural framework for continuous optimization through their infinite-dimensional Hilbert spaces. In this paper, we propose the Complex Continuous-Variable Quantum Approximate Optimization Algorithm (CCV-QAOA), a variational framework operating in the complex domain that optimizes over complex decision variables. The method efficiently solves real and complex multivariate optimization problems. To demonstrate its versatility, we apply CCV-QAOA across a broad suite of optimization use cases, including convex quadratic minimization, scaling studies with circuit depth and cutoff dimension, constrained quadratic programs using penalty constructions, and non-convex benchmarks such as the Styblinski-Tang function and complex quartic landscapes
Propagation and scattering of ultrasonic waves in macroscopically anisotropic polycrystalline materials with fiber texture
No full textInternational audienceThis work investigates the scattering of ultrasonic elastic waves in polycrystalline materials with macroscopic transverse isotropy (TI). Using the Dyson equation with the First-order smoothing approximation (FOSA) for the mass operator, we obtain the complex wavenumber of the qL, T 1 , and qT 2 waves in 3D and qL and qT 2 in 2D. The equations are properly modified by including the Green solution for the reference TI homogeneous medium. Additionally, 2D numerical finite element (FE) models are developed, allowing for a direct comparison with the 2D theoretical amplitude attenuation and phase velocities estimations. The numerical simulations are carried out with a set of periodic boundary conditions (PBC), leading to a proper plane wave propagation even in the case of macroscopic anisotropy. The influence of the orientation of the axis of symmetry on the attenuation and phase velocities is discussed. The role of the symmetry-axis orientation is analyzed, and scattering mechanisms are contrasted between 2D and 3D. Numerical and theoretical results show excellent agreement, confirming the increased wave dispersion when propagation occurs along the cross-fiber (or cross-symmetry-axis) direction
Dual-Polarized Array of Triangular Leaky-Wave Antennas for Compact Water Radar Instruments
No full textNational audienceThis paper presents a compact dual-polarized leaky-wave antenna array for water remote sensing. The antenna employs a triangular slotted waveguide to achieve high gain and slanted pointing. Its cross-section supports two orthogonal propagation modes to enable dual-linear polarization and a highly slanted beam without exciting higher-order modes. A tailored slot shape is introduced to provide high leakage for both modes while efficiently controlling the aperture illumination to reduce sidelobe levels. The antenna consists of a three-layer monolithic structure, where the array of triangular leaky waveguides supports two stacked beamforming networks (one per polarization). An array of orthomode transducers connects the beamformers to the radiating elements. Unconventional waveguide sections are employed to ensure self-supporting additive manufacturing. The design is validated through a progressive methodology that increases complexity in steps, allowing separate validation of key features and incorporating feedback from the manufacturing process to minimize deviations. Both the antenna concept and the experimental methodology are confirmed through prototyping of a 4-element array and a 32-element array. The final design achieves realized gains above 37 dB, high cross-polarization discrimination and isolation, and reduced sidelobe levels. Moreover, high slanted angles are attained without generating grating lobes or exciting higher-order modes, meeting the specifications of an airborne water monitoring radar instrument
Data-Driven Joint Complex-Valued Chance-Constrained Programs under Wasserstein Ambiguity Set
No full textIn this paper, we study data-driven chance-constrained optimization with complexvalued decision variables (3CP), motivated by complex-valued linear programs arising in signal processing and communications. To hedge against distributional uncertainty in the random coefficients, we adopt a distributionally robust formulation over a Wasserstein ambiguity set centered at the empirical distribution. For both individual and joint probabilistic constraints, we derive tractable approximations via worst-case Conditional Value-at-Risk (CVaR) reformulations. The individual constraints lead to a convex, conservative, and computationally efficient program, whereas the joint constraints yield a biconvex approximation. To address the latter, we develop a convex sequential method that produces feasible upper bounds, and we embed it within a spatial branch-and-bound framework that leverages monotonic relaxations to compute certified lower and upper bounds. Numerical experiments confirm the effectiveness of the approach: the convex sequential method stabilizes within a few iterations, and the branch-and-bound optimality gap shrinks and converges to zero (up to a prescribed tolerance) as the search tree is refined.</div
Extended formulations for induced tree and path polytopes of chordal graphs
No full textInternational audienceExtended formulations for induced tree and path polytopes of chordal graph
Transient power unit efficiency prediction of a small data center through thermal and energetic analyses
No full textInternational audienceData Centers (DCs) are proliferating throughout the world, and their energy consumption is increasing. Therefore, optimizing the energy efficiency of the server environment becomes crucial. For this purpose, the usual key metric is the Power Unit Efficiency (PUE). However, this measurement only occurs when a Data Center (DC) becomes operational. The current study builds a methodological tool to estimate PUE~ metrics before the DC is built, by bringing together concepts from Information Technology (IT) and energy laws. Firstly, a consumption study is carried out on a small university DC with a power of 140 kW. An energy behavior pattern is drawn based on 5 minutes time-step data collected over a recent period of 7 years, from 2016 to 2023, making this dataset relevant for a benchmark of historical DC server energy consumption. Secondly, a dynamic model is built to integrate the variable energy behavior and the weather-dependent system environment. In this way, the short time PUEτ reflects interesting non-constant values, exhibiting maximum and minimum performances of the DC. The results of the dynamic model are compared to the actual measurements and show a deviation of 5% between measurements and predictions. Temporal variations of PUE are strongly correlated with temperature variations during a day. In contrast, server environments exhibit an almost constant consumption, except for High Performance Computing (HPC) during the working days, for which the energy demand is higher than during nights. This approach is also used to make long-term predictions evaluating the impact of global warming on DC efficiency as an example. In this study, a prediction was made for the years 2035 and 2065 and could be easily implemented by manufacturers as a reliable method to estimate PUE~ during the pre-construction and projection phases of DC in a context of rapid climate change
High-speed line-field confocal optical coherence tomography
No full textInternational audienceLine-field confocal optical coherence tomography (LC-OCT) is an optical imaging technique based on low-coherence interference microscopy that uses line illumination and line detection to generate tomographic images of semi-transparent samples with micrometer-scale spatial resolution. We present an LC-OCT system that achieves a threefold increase in acquisition speed compared with previously reported systems, without compromising image quality in terms of spatial resolution, contrast, or field of view. Through modifications to the experimental set-up, as well as changes and optimization of image acquisition and processing, the system captures vertical section images (1250 μm × 500 μm, lateral (x) × axial (z)) and horizontal section images (1250 μm × 500 μm, lateral (x) × lateral (y)) at a frame rate of 30 Hz. High speed in vivo imaging of human skin is demonstrated with a resolution of 1.4 μm × 1.1 μm (lateral × axial). No motion blur is observed during rapid scanning of large skin areas. Notably, the system allows visualization of capillary blood flow by tracking the motion of individual red blood cells