HAL-Université de Bretagne Occidentale
Not a member yet
75442 research outputs found
Sort by
LiteInjector: A LiteX Extension for Fault Injection
No full textInternational audienceIn this article, we focus on the emulation of fault injections on FPGA boards using saboteurs by presenting LiteInjector.LiteInjector is an open source bit- and cycle-accurate logic fault emulator written in Python, developed with the aim of accelerating security evaluation campaigns for systems-on-chip (SoC). LiteInjector has been tested on several use cases. These use cases rely on VerifyPin codes from the FISSC security benchmark running on a Linux-capable system-on-chip with a RISC-V core. We demonstrate that the tool provides a flexible solution that supports a large set of fault models. Furthermore, results show that, compared to an HDL simulation-based tool, LiteInjector allows reducing the fault injection campaign time by a factor of 210 considering a Linux-based system
Shoulder Endurance Test : étude monocentrique prospective exploratoire interventionnelle et randomisée chez des handballeuses de 14 à 18 ans en structure intensive d'entraînement
No full textGeneral context : Shoulder injuries are common in handball, particularly among young female athletes. The Shoulder Endurance Test is designed to assess shoulder muscle endurance and could play a role in injury prevention. However, this test has not yet been studied in this specific population. Establishing normative values for this test could help identify risk factors and develop injury prevention strategies. Methods : This is a prospective, exploratory, interventional, randomised, single-centre study of 14-18 year old female handball players in an intensive training structure. The test was performed on both the dominant and non-dominant arms to obtain reference values. Statistical analyses were conducted to explore potential associations between test performance and factors such as history of injury, playing position, and age. Results : Preliminary findings suggest that factors such as previous shoulder injuries and pre-test fatigue levels may influence test outcomes. The study aims to establish normative values (dominant arm : 81,8 ; non dominant arm : 71), which will allow for better interpretation of test results in a handball-specific context. Additional statistical analyses were used to determine whether statistically significant differences were observed. Playing position, age, fatigue and history of shoulder injuries were analysed. Conclusion : This study provides initial reference values for the Shoulder Endurance Test in female handball players. However, due to the limited sample size, these results should be interpreted with caution. Future research with a larger cohort is necessary to confirm these findings.Introduction : Les blessures à l'épaule sont fréquentes dans le handball, en particulier chez les jeunes athlètes féminines. Le Shoulder Endurance Test est conçu pour évaluer l'endurance des muscles de l'épaule. Cependant, ce test n'a pas encore été étudié dans cette population spécifique. Établir des valeurs normatives pour ce test pourrait aider à identifier les facteurs de risque et à développer des stratégies de prévention des blessures. Méthode : Il s'agit d'une étude prospective, exploratoire, interventionnelle, randomisée et monocentrique portant sur des joueuses de handball âgées de 14 à 18 ans dans une structure intensives d'entraînement. Le test a été réalisé sur le bras dominant et sur le bras non dominant afin d'obtenir des valeurs de référence. Des analyses statistiques ont été menées pour explorer les associations potentielles entre la performance du test et des facteurs tels que les antécédents de blessure, la position de jeu et l'âge. Résultat : Les résultats préliminaires suggèrent que des facteurs tels que les blessures antérieures à l'épaule et la fatigue avant le test peuvent influencer les résultats du test. L'étude vise à établir des valeurs normatives (bras dominant : 81,8 ; bras non dominant : 71). Des analyses statistiques supplémentaires ont été utilisées pour déterminer si des différences statistiquement significatives ont été observées. La position de jeu, l'âge, la fatigue et les antécédents de blessures à l'épaule ont été analysés. Conclusion : Cette étude fournit des valeurs de référence initiales pour le test SET chez les joueuses de handball. Cependant, en raison de la taille limitée de l'échantillon, ces résultats doivent être interprétés avec prudence. Des recherches futures avec une plus grande cohorte sont nécessaires pour confirmer ces résultats et améliorer la fiabilité du test pour la prévention des blessures dans le handball
Ocean Literacy as a Mechanism for Change Across and Beyond the UN Ocean Decade
No full textInternational audienceUnderstanding the complexity of human-ocean relationships has been increasingly recognized as being central to addressing the triple crises currently facing the ocean and the communities that depend on it-climate change, biodiversity loss, and social inequities. Since the early 2000s, the concept of ocean literacy (OL) has evolved as a framework to explore and critically assess this relationship. Defined as having an understanding of "your influence" on the ocean and its "influence on you," OL has moved beyond its original education and knowledge roots to recognize at least 10 dimensions. These dimensions-which include themes of knowledge, emotions, attitudes and communication, and the frameworks associated with them-are increasingly being adopted, and indeed adapted, to help to further understand human-ocean relationships and to support the co-development of solutions to address ocean challenges. With the positioning of OL as a key mechanism for change within the UN Decade of Ocean Science for Sustainable Development, it is both timely and necessary to continue to examine and expand the concept of OL beyond its existing boundaries. This thematic issue demonstrates the interdisciplinarity of OL research, presenting a range of studies that critically explore how the dimensions, drivers, and impacts of OL can vary in different socio-cultural, economic, political, and geographic contexts. These studies provide crucial insight into the developing role of OL within wider ocean governance and sustainability processes. Collectively, the articles highlight the diversity of ocean literacy research emerging from the community, insights into how to further develop OL initiatives and how to maximise the potential of OL as a mechansim for change across the Ocean Decade and beyond
Navigating the Digital Twin Landscape: Aligning Multiplicity of Tools with Multi-Use Case Systems
No full textInternational audienceDigital twins anchor Industry 4.0, enabling simulation, analysis, and optimization of physical systems through virtual counterparts. While benefits span healthcare, city planning, buildings, energy, education, and smart manufacturing, organizations often misalign solutions with user expectations. Key pitfalls include unclear requirements, overengineered platforms, and interoperability struggles. This paper consolidates insights from the literature and underlines the importance of early stakeholder involvement, well-defined performance indicators, and deferring technological tool selection until the final stages of the design process. By choosing platforms and methods only after user needs are clearly articulated and validated, digital twin implementations can better match operational realities. Case studies in didactic production lines, agro-industrial bioreactors, and short-distance food supply chains confirm how a usercentric, adaptive, and structured approach fosters more scalable, relevant, and sustainable outcomes
Building a MIDI-Compatible Subtractive Synthesizer on STM32 for Embedded Systems & Signal Processing Teaching
No full textInternational audienceThis article presents an STM32-based project designed for final-year GEII BUT students in Electronics and Embedded Systems at the IUT of Brest-Morlaix. The objective of this project is to encourage students to integrate multiple skills in signal processing, electronics, embedded systems, and C programming during a 20-hour project. The project is implemented on an STM32 Discovery kit and is divided into two main phases: the first 16 hours are dedicated to developing the fundamental building blocks of a digital sound synthesizer (digital oscillators with interpolated lookup wavetables, parametric digital IIR filter, modulators, and MIDI over USB connection). During the remaining four hours, students design and implement their own synthesizer concepts based on their interests, under the supervision and guidance of the instructor. Student feedback highlights the project's effectiveness in bridging theoretical knowledge and hands-on implementation
Contrôler et garantir l'usage de la force étatique.Le cadre réglementaire des lanceurs de balles de défense
No full textNational audienceLes lanceurs de balles de défense ont fait l’objet d’une réglementation importante encadrant leur emploi et leurs précautions d’usage. L’étude de ces textes renseigne sur la façon dont les autorités hiérarchiques de la police nationale et de la gendarmerie réglementent l’usage de la force par l’État mais aussi sur la manière dont elles se représentent l’intensité acceptable de la contrainte physique armée contre les administrés. Ainsi, encadrer l’usage des lanceurs de balles de défense permet, en restreignant et en contrôlant l’usage de la force étatique, d’en garantir le principe
Assessing Local Inductive Bias in Deep Learning-Based Three-Gamma Event Classification
No full textInternational audienceIn the context of the Xenon Medical Imaging System (XEMIS) project-developing the first liquid xenon (LXe) Compton camera for functional three-gamma medical imaging-we explore the application of deep learning methods for event classification based on the number of observed gamma-rays per event. Accurately identifying the number of gamma rays interacting with the liquid xenon (LXe) target per event is crucial for enabling event-byevent targeted data treatment and image reconstruction techniques. To this end, we evaluated the classification accuracy of two deep learning models with or without local inductive bias (Sparse Tensor Network and PointNet). Both models process unstructured raw spatial and light detection data from Geant4 simulated events, aiming to distinguish between single-, double-, and triple-gamma ray interactions. Our results offer insights into the feasibility and performance of sparse tensor architectures in handling complex data typical of Compton imaging systems.</div
Density-based isolation forest
No full textInternational audienceThis work addresses the problem of outlier detection and studies isolation forest algorithms, originally introduced by Liu et al. (2008). We propose a new algorithm, which we call the density-based isolation forest (dbiForest). A dbiForest consists of a large number of density-based isolation trees (dbiTrees). Each dbiTree aims to identify outliers through recursive binary splits based on multivariate density estimates. This original approach integrates density estimation into the structure of the isolation tree. Based on the assumption that outliers are both rare and different from normal values, this algorithm identifies outliers as points that lie in regions of low density. Through several numerical experiments, we empirically show that this new method can overcome some of the limitations of existing isolation forest algorithms.</div
