Portail HAL Ensta
Not a member yet
11080 research outputs found
Sort by
Stochastic incremental mirror descent algorithms with Nesterov smoothing
No full textInternational audienceFor minimizing a sum of finitely many proper, convex and lower semicontinuous functions over a nonempty closed convex set in an Euclidean space we propose a stochastic incremental mirror descent algorithm constructed by means of the Nesterov smoothing. Further we modify the algorithm in order to minimize over a nonempty closed convex set in an Euclidean space a sum of finitely many proper, convex and lower semicontinuous functions composed with linear operators. Next a stochastic incremental mirror descent Bregman-proximal scheme with Nesterov smoothing is proposed in order to minimize over a nonempty closed convex set in an Euclidean space a sum of finitely many proper, convex and lower semicontinuous functions and a prox-friendly proper, convex and lower semicontinuous function. Different to the previous contributions from the literature on mirror descent methods for minimizing sums of functions, we do not require these to be (Lipschitz) continuous or differentiable. Applications in Logistics, Tomography and Machine Learning modelled as optimization problems illustrate the theoretical achievements
A multiscale high cycle fatigue approach based on shakedown and cohesive zone theories—Application for additive-manufactured 316L steel
No full textInternational audienc
Titanium-mediated expedient synthesis of complex nitrogen heterocycles
No full textInternational audienc
Highlights from the 55th Bürgenstock Conference on Stereochemistry 2022
No full textInternational audienceIn May 2022, the 55th Bürgenstock Conference on Stereochemistry happened in person once again. This summary provides insight into the scientific themes discussed during the most recent meeting of this historic and multi-disciplinary conference
Fatigue life of carbon-epoxy laminated composites: combining thermo-mechanical analysis and residual strength model
No full textInternational audienceLong-term durability under repeated mechanical loading is a major criterion for design of composite structures. The development of a fatigue criterion faces two major difficulties: (i) the proposal of a load/life curve (S-N curve) describing the natural dispersion of the fatigue tests and (ii) the duration of these tests. In order to deal with the first point, Sendeckyj proposed an approach based on a so-called residual strength that fits S-N curve and associates a failure probability [1]. The procedure to identify this model requires quasi-static and fatigue tests. This fatigue campaign could be large and time consuming. A solution to decrease the duration of the tests consists in using a self-heating-based method [2]. The principle is to estimate a mean S-N curve from the thermal dissipation, inferred from temperature measurements on a single sample. This method reduces drastically the duration of the fatigue tests campaign. However, neither the associated dispersion nor the low cycle fatigue part is estimated for now for this class of materials. The goal of the study is to combine the two approaches (residual strength and thermal measurements) in order to identify the S-N curve and the related dispersion. The fatigue lifetime for high cycles loading is estimated by the heat-build up protocol. Quasi-static tests and a few fatigue tests are used to estimate the failure distribution based on a two-parameters Weibull model. The article will detail this approach and will present the results obtained on laminated composites carbon-epoxy manufactured from unidirectional plies and 2D-woven plies
From Modeling to Sensing of Micro-Doppler in Radio Communications
No full textInternational audienceThe Doppler effect in radio systems has been widely explored by the radio communication community. However, these studies have been limited to simple motion such as linear translation. This paper presents a model for the Doppler modulation effect, i.e., the effect of complex movement on the received signal, using a geometrical approach. Particularly, we focused on studying micro-Doppler in radio communications produced by vibrations. Exploiting this phenomenon would allow the performance of passive micro-Doppler effect sensing based on communication. In this paper, we also propose signal processing techniques to detect the presence of the micro-Doppler effect and to estimate its parameters. Then, we present some experiments which highlight the micro-Doppler effect in a radio communication context. Finally, the end of the paper discusses some potential applications that exploit this phenomenon
Seismic analysis of tall buildings through an enriched equivalent beam model: Application to Grenoble City Hall
No full textInternational audienceThe seismic assessment of multi-story buildings at city scale requires robust but simplified modeling techniques to reduce the computational resources and time. This paper presents an equivalent beam model (EBM), issued from homogenization, capable of describing the seismic response of real tall buildings. The EBM is a generalization of the Timoshenko beam model. It integrates shear, global bending, and inner bending mechanisms, which are identified to control the response of reticulated structures such as buildings, foams, honeycombs, or lattice structures. The parameters of the EBM are obtained through the static analysis of a single building story. This work uses a new finite element formulation of the EBM to estimate the dynamic properties of buildings (i.e., eigenfrequencies and eigenmodes) and perform time history analyses with natural seismic records. This approach is validated through a real case study. The Grenoble City Hall corresponds to a regular 13-story reinforced concrete building located in Grenoble-France. The inner bending significantly influences its behavior, whereas usual beam models do not include this mechanism. The results obtained via this 1D simplified EBM match those obtained with a fully 3D finite element simulation, which are also very close to experimental data
A Two-Way Split-Step Wavelet Scheme for Tropospheric Long-Range Propagation in Various Environments
No full textInternational audienceIn the context of improving the dimensioning of observation and telecommunication, the characterization of the propagation canal is very important. Thus, accurate models of propagation phenomenona in their environment and above a rough surface (maritime or terrestrial) are of major interest for many applications (such as radar, communications, and teledetection). To provide solutions to this problem, in this paper, we propose a fast, memory-efficient, and accurate asymptotic method for 2D tropospheric propagation for a large band of frequency that accounts for relief, as well as ground composition and roughness. This latter is a two-way split-step wavelet scheme with an intrinsic stopping criterion. For overseas propagation, roughness effects are considered through a hybrid method. A complete theoretical comparison with SSF in terms of memory and time efficiency is proposed. Simulations in various environments (ground, sea, and snow), as well as different frequencies (UHF, S, and X-band) are performed to validate the method and highlight its advantages. To highlight the interest of the developed methodology, this latter is applied to different real-life applications, such as the prediction of radar coverage and the optimization of an antenna location
Protecting Behavioral IPs During Design Time: Key-Based Obfuscation Techniques for HLS in the Cloud
No full textInternational audienceGlobalization of the IC supply chain and the ever more distributed character of hardware design flows have lead to a surge in security threats. While the focus has long been on IP theft and counterfeiting issues during manufacturing, concern about vulnerabilities at design time has been growing more prevalent in recent years. With the advent of new cloud-based business-to-business solutions, cloud-based EDA services are likely to gain widespread adoption in the future. However, such services can increase the risks of theft. In this chapter, we focus on the scenario of a cloud-based HLS service and on how to protect Behavioral Intellectual Property (BIP) against theft at design time. We show how software obfuscation techniques can be adapted to this hardware context. By adding a de-obfuscation step after HLS, and reusing hardware logic locking principles, we propose a transient, key-based obfuscation scheme to protect BIPs during HLS with low design overhead. We illustrate this scheme with a presentation and study of several practical obfuscation techniques
MUAD: Multiple Uncertainties for Autonomous Driving, a benchmark for multiple uncertainty types and tasks
No full textAccepted at BMVC 2022International audiencePredictive uncertainty estimation is essential for safe deployment of Deep Neural Networks in real-world autonomous systems. However, disentangling the different types and sources of uncertainty is non trivial for most datasets, especially since there is no ground truth for uncertainty. In addition, while adverse weather conditions of varying intensities can disrupt neural network predictions, they are usually under-represented in both training and test sets in public datasets.We attempt to mitigate these setbacks and introduce the MUAD dataset (Multiple Uncertainties for Autonomous Driving), consisting of 10,413 realistic synthetic images with diverse adverse weather conditions (night, fog, rain, snow), out-of-distribution objects, and annotations for semantic segmentation, depth estimation, object, and instance detection. MUAD allows to better assess the impact of different sources of uncertainty on model performance. We conduct a thorough experimental study of this impact on several baseline Deep Neural Networks across multiple tasks, and release our dataset to allow researchers to benchmark their algorithm methodically in adverse conditions. More visualizations and the download link for MUAD are available at https://muad-dataset.github.io/