Search

International audience; Digital signal processing for fiber nonlinearity com- pensation is a key enabler for the ever-increasing demand for higher data rates in coherent optical transmissions. A major challenge of existing techniques is that the fiber nonlinear coefficient needs to be scaled properly during compensation in order to reach the achievable signal quality increase. We solve this problem using a low-complexity algorithm adaptively optimizing a metric based on the soft-decision bitwise demodulator used for modern FEC decoders. An analytical model shows that the proposed scheme converges to the optimal scaling factor with a predictable precision, that is validated by numerical results

Global data traffic is expected to reach up to 4.3 ZB per year by 2020. With the majority of the global communications being transported on submarine point-to-point fiber-optic systems, different cutting-edge technologies have been under research to cope with this unprecedented traffic growth. Continuous advances in high-speed integrated circuits have allowed the use of advanced modulation formats and digital signal processing (DSP) techniques to maximize the transmission spectral efficiency. With mitigation of fiber linear effects efficiently carried out by DSP with relative low-complexity, the capacity of modern fiber optic systems rests limited by fiber nonlinearities. To this extent, in the first part of this work, the performance and achievable benefits of low-complexity DSP techniques aiming to mitigate fiber Kerr nonlinear effects are investigated. Besides nonlinear compensation techniques, the use of multi-level modulation formats beyond 16QAM and high symbol rate channels have gained momentum to increase the system spectral efficiency. One of the major breakthroughs in the recent years, has been the introduction of QAM-based probabilistic constellation shaping (PCS-QAM), which has proven to outperform regular QAM formats. In this sense, in the second part of this work, the practical achievable rate increase brought by PCS-QAM for transoceanic distances is investigated. A theoretical and experimental comparison with other high-capacity formats is performed, and the design of a PCS-QAM for trans-Pacific distances is addressed. Finally, in the last section, several transmission records using the two above techniques are reported, La majeure partie des communications mondiales est transportée par des systèmes transocéaniques à fibre optique. Il est estimé que d'ici 2020 le trafic de données atteindra 4.3 ZB par an. Afin de faire face à cette demande, différentes technologies sont actuellement étudiées pour augmenter la capacité des systèmes de transmission très longue distance. Avec l'avènement des circuits intégrés à haute vitesse, des formats de modulation avancés et des techniques de traitement de signal numérique (DSP) peuvent être utilisés pour maximiser l'efficacité spectrale de transmission. Par ailleurs, la capacité des systèmes modernes est fortement limitée par les effets non-linéaires de type Kerr dans la fibre. Ainsi, la première partie de ce travail est axée sur l’étude de la performance et des gains réalisables par des techniques DSP à faible complexité pour mitiger les effets non-linéaires monocanal. En outre, l’utilisation des formats de modulation multiniveaux à haute efficacité spectrale au-delà de 16QAM a pris de l'ampleur pour augmenter le débit de transmission des systèmes, notamment avec l’introduction des formats QAM avec mise en forme probabiliste (PCS-QAM), plus performants que les formats QAM classiques. La deuxième partie de ce travail présente donc une comparaison théorique ainsi qu’expérimentale du format PCS-64QAM avec d’autres formats à haute efficacité spectrale pour les distances transatlantiques. La mise en œuvre d’un format PCS-64QAM conçu pour les distances transpacifiques est également abordée. Enfin, la dernière partie de ce travail concrétise les résultats des travaux menés dans les deux sections précédentes en présentant plusieurs records de transmission

International audience; This study addresses the problem of Alzheimer's disease (AD) diagnosis with Electroencephalography (EEG). The use of EEG as a tool for AD diagnosis has been widely studied by comparing EEG signals of AD patients only to those of healthy subjects. By contrast, we perform automated EEG diagnosis in a differential diagnosis context using a new database, acquired in clinical conditions, which contains EEG data of 169 patients: subjective cognitive impairment (SCI) patients, mild cognitive impairment (MCI) patients, possible Alzheimer's disease (AD) patients, and patients with other pathologies. We show that two EEG features, namely epoch-based entropy (a measure of signal complexity) and bump modeling (a measure of synchrony) are sufficient for efficient discrimination between these groups. We studied the performance of our methodology for the automatic discrimination of possible AD patients from SCI patients and from patients with MCI or other pathologies. A classification accuracy of 91.6% (specificity = 100%, sensitivity = 87.8%) was obtained when discriminating SCI patients from possible AD patients and 81.8% to 88.8% accuracy was obtained for the 3-class classification of SCI, possible AD and other patient

International audience; We extract voice features from read speech of patients diagnosed as having Alzheimer's disease and mild cognitive impairment versus healthy control group at the AP-HP Broca Hospital. Two classification methods are used to evaluate the efficiency of these features. Finally, a wrapper feature selection method as exploits both methods is also evaluated

International audience; In this paper we investigated two solutions for distributing 802.15.3c signal over Multi-Mode Fiber (MMF) to solve the problem of limited coverage at 60 GHz. The first solution consists of the transmission of the signal at 60 GHz directly over fiber (RFoF) and the second solution consists of the transmission of the signal at intermediate frequency 5 GHz over fiber (IFoF). The millimeter wave is generated using the Optical Carrier Suppression (OSC) technique, and the optical link is based on MMF without optical amplifiers in order to obtain a low cost architecture. The systems are evaluated and analyzed by Bit Error Rate (BER) and Error Vector Magnitude (EVM) performance. The comparison between the two solutions is based on the fiber length, the complexity and cost of the implementation

National audience; Nous étudions expérimentalement une solution hybride WDM/DS-OCDMA pour l'application réseau d'accès optique haut débit. Cette solution permet d'augmenter la capacité de multiplexage du réseau WDM en y ajoutant une dimension supplémentaire : le code temporel. Le système est réalisé dans une configuration mono-canal WDM. Les performances de la transmission, en termes de taux d'erreurs binaires, sont évaluées après 20 km de transmission.

International audience; Intended for the next generation of optical access networks, OCDMA is of great interest to meet the demand of increasing the number of users per access fiber, especially as spectral phase coding increases its performance in the optical domain. This, however, requires handling broad spectra and short pulses, which are best dealt with using opto-electronic or all-optical devices instead of slower electronics. Among others, we demonstrate spectral-phase-coded OCDMA using a fiber-based saturable absorber as thresholding in the receiver.

International audience

National audience

National audience; Nous présentons une solution hybride WDM/DS-OCDMA pour l'application réseau d'accès optique haut-débit.

International audience; Nous étudions la génération d'un supercontinuum cohérent dans une fibre optique non-linéaire à dispersion normale aux longueurs d'onde des télécommunications optiques.

International audience

International audience; Single section quantum dash (QDash) mode locked lasers (MLL) can provide a flat and broadband optical frequency comb with low energy consumption, operational simplicity and large-scale low-cost production possibilities. MLL longitudinal modes can be employed as single carriers with a regular spectral spacing in a dense wavelength division multiplexing (DWDM) link, making them promising components for next generation DWDM transceivers. However, individual modes of a MLL suffer from relatively high phase noise while high capacity coherent transmission requires carriers with low phase noise. Optical feedback, which is a well-known method to reduce the linewidth of a single mode laser, can be used to stabilize comb lasers. This paper reports on the investigation of phase noise properties of comb lines delivered from a single section QDash MLL under optical feedback and shows that each MLL longitudinal mode optical linewidth can be drastically narrowed. It enables coherent transmission with extended link budget over 50 km at Tbit/s line rates.

International audience; We report on the use of semiconductor optical amplifiers (SOAs) to extend the optical bandwidth of next generation optical systems to 100 nm and beyond. After discussing the technological progress and the motivation for rekindling the interest in SOAs for line amplification, we describe the innovative approach developed for the realization of ultra-wideband (UWB) SOAs. Leveraging custom design of singly polarized SOAs to provide gain over 100+ nm bandwidth, we developed a polarization diversity architecture to realize UWB-SOA modules. Embedded in a compact package, the UWB amplifier modules have been successfully used to demonstrate 100+ Tb/s transmissions. We subsequently review recent experimental transmission results based on such novel 100+ nm wide semiconductor optical amplifiers, including our first demonstration of 100+Tb/s transmission over 100 km distance, our field trial using real-time traffic, and finally the transmission of 107 Tb/s throughput over three spans of standard single mode fiber (SSMF) using hybrid UWB Raman/SOA amplification technique.

A new multimodal biometric database designed and acquired within the framework of the European BioSecure Network of Excellence is presented. It is comprised of more than 600 individuals acquired simultaneously in three scenarios: 1) over the Internet, 2) in an office environment with desktop PC, and 3) in indoor/outdoor environments with mobile portable hardware. The three scenarios include a common part of audio/video data. Also, signature and fingerprint data have been acquired both with desktop PC and mobile portable hardware. Additionally, hand and iris data were acquired in the second scenario using desktop PC. Acquisition has been conducted by 11 European institutions. Additional features of the BioSecure Multimodal Database (BMDB) are: two acquisition sessions, several sensors in certain modalities, balanced gender and age distributions, multimodal realistic scenarios with simple and quick tasks per modality, cross-European diversity, availability of demographic data, and compatibility with other multimodal databases. The novel acquisition conditions of the BMDB allow us to perform new challenging research and evaluation of either monomodal or multimodal biometric systems, as in the recent BioSecure Multimodal Evaluation campaign. A description of this campaign including baseline results of individual modalities from the new database is also given. The database is expected to be available for research purposes through the BioSecure Association during 2008, Published at IEEE Transactions on Pattern Analysis and Machine Intelligence journal

International audience; We present a control-theoretic approach to achieve Cyber-Physical Systems (CPS) resilient designs. We assume situations in which the CPS must maintain the correct operation of a set of crucial functionalities despite ongoing adversarial misbehavior. The approach is based on a moving target defense paradigm, driven by a linear switching of state-space matrices, and applied at both the physical and network layers of a networked-control system. We show that the final system maintains stability. We also evaluate, via simulation, a step-by-step procedure that takes a transfer function, representing the dynamics of the physical process. As a result, we obtain a resilient CPS design structured around a topology of decentralized controllers. Results show that the obtained approach is both innovative and promising.

International audience; We describe the bifurcation route of a semiconductor laser with external optical feedback by use of Lyapunov exponent/fractal. The Feigenbaum «r∞-value» is proposed as the critical value for identifying the limit of the Lang-Kobayashi model.

International audience; We investigate theoretically the stability properties of semiconductor lasers with optoelectronic feedback created by external optical feedback, by means of asymptotic methods. The impact of this dual-feedback configuration on the first Hopf bifurcation is discussed.

International audience; Indoor air quality (IAQ) is a growing and multidisciplinary research field domain involving professionals with various expertise, from Materials Sciences, Instrumentation, Electronics, Physics, Chemistry and many more. One of the current hot topics of this field is to equip private accommodation with air quality monitors. For that, it is crucial to develop and choose the more sui table sensors capable of detecting and discriminate different types of gases and volatile organic compounds. The number of needed sensors is related to which substances are going to be monitored. However, this choice is not obvious as the intrinsic information of sensors is not always well understood. The main objective of this work is to understand and provide a framework on how this choice can be made. Several different types of signals are generated when using multisensory systems, but not all the signals are useful for detecting the aimed substances. By analyzing these signals, it is possible to select the sensors presenting the best match with a given substance detection task. To analyze the collected data, a technique of intrinsic dimension estimation was used. This technique allows the estimation of how many dimensions (independent variables) are necessary to synthesize equivalent signals. In the case of multisensory systems, the dimensions can be interpreted as the number of sensors present in it. However, the intrinsic dimension estimation of a dataset collected via multisensory systems may reveal a smaller effective dimension, thus possibly allowing the estimation of the minimum number of sensors needed for a given task. The dataset analyzed in this work is composed by the responses of 40 RUBIX-PODs with 19 sensors each, during 6 experiments where specific gas injections are done to represent different indoor activities (cleaning, cooking, etc.). The intrinsic dimension (ID) of this dataset is estimated. The data was arranged accordingly to an injection discrimination task, i.e., detecting what experiment (here called injection) generated each data point. Preliminary results showed an ID of 3 for the given dataset and indeed when only three specific sensors were used, all the injections could visibly be separated. Although this result shows that ID estimation is a good indication of the number of sensors needed, this experiment is very simple compared to what the multisensory systems face during day-to-day use, more experiments using real world data collected by those systems should provide a better understanding on what they are detecting and what sensors are really needed.

International audience; The elderly living at home are in most of the cases (concerning Western and Central Europe and North America) living alone and with an increased risk of accidents. In France, about 4.5 % of men and 8.9% of women aged of 65+ years has an everyday life accident . Between these everyday life accidents, the most important part is represented by the domestic accidents; about 61% (same source) and 54% of them take place inside the house. In France, annually, 2 millions of elderly falls take place, which represent the source of 10 000 deaths . Between 30% and 55% of falls cause bruises and only 3% to 13% of falls are the causes of serious injuries such as fractures, dislocation of a joint, or wounds. Apart from physical injury and hospitalization, a fall can cause a shock (especially if the person cannot recover only after the fall). This condition can seriously affect the senior psychology, he might looses the confidence in his abilities and can result in a limitation of daily activities and, consequently, in a decrease of the life quality. In order to improve the quality of life of elderly several applications has been developed: home telemonitoring in order to detect distress situations and audio-video transmission in order to allow specialists to diagnose patient at distance. This chapter describe a medical remote monitoring solution allowing the elderly people to live at home in safety

Many articles have used voice analysis to detect Parkinson's disease (PD), but few have focused on the early stages of the disease and the gender effect. In this article, we have adapted the latest speaker recognition system, called x-vectors, in order to detect an early stage of PD from voice analysis. X-vectors are embeddings extracted from a deep neural network, which provide robust speaker representations and improve speaker recognition when large amounts of training data are used. Our goal was to assess whether, in the context of early PD detection, this technique would outperform the more standard classifier MFCC-GMM (Mel-Frequency Cepstral Coefficients - Gaussian Mixture Model) and, if so, under which conditions. We recorded 221 French speakers (including recently diagnosed PD subjects and healthy controls) with a high-quality microphone and with their own telephone. Men and women were analyzed separately in order to have more precise models and to assess a possible gender effect. Several experimental and methodological aspects were tested in order to analyze their impacts on classification performance. We assessed the impact of audio segment duration, data augmentation, type of dataset used for the neural network training, kind of speech tasks, and back-end analyses. X-vectors technique provided better classification performances than MFCC-GMM for text-independent tasks, and seemed to be particularly suited for the early detection of PD in women (7 to 15% improvement). This result was observed for both recording types (high-quality microphone and telephone)., Submitted

Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Automatically verifying the identity of a person by means of biometrics (e.g., face and fingerprint) is an important application in our day-to-day activities such as accessing banking services and security control in airports. To increase the system reliability, several biometric devices are often used. Such a combined system is known as a multimodal biometric system. This paper reports a benchmarking study carried out within the framework of the BioSecure DS2 (Access Control) evaluation campaign organized by the University of Surrey, involving face, fingerprint, and iris biometrics for person authentication, targeting the application of physical access control in a medium-size establishment with some 500 persons. While multimodal biometrics is a well-investigated subject in the literature, there exists no benchmark for a fusion algorithm comparison. Working towards this goal, we designed two sets of experiments: quality-dependent and cost-sensitive evaluation. The quality-dependent evaluation aims at assessing how well fusion algorithms can perform under changing quality of raw biometric images principally due to change of devices. The cost-sensitive evaluation, on the other hand, investigates how well a fusion algorithm can perform given restricted computation and in the presence of software and hardware failures, resulting in errors such as failure-to-acquire and failure-to-match. Since multiple capturing devices are available, a fusion algorithm should be able to handle this nonideal but nevertheless realistic scenario. In both evaluations, each fusion algorithm is provided with scores from each biometric comparison subsystem as well as the quality measures of both the template and the query data. The response to the call of the evaluation campaign proved very encouraging, with the submission of 22 fusion systems. To the best of our knowledge, this campaign is the first attempt to benchmark quality-based multimodal fusion algorithms. In the presence of changing - - image quality which may be due to a change of acquisition devices and/or device capturing configurations, we observe that the top performing fusion algorithms are those that exploit automatically derived quality measurements. Our evaluation also suggests that while using all the available biometric sensors can definitely increase the fusion performance, this comes at the expense of increased cost in terms of acquisition time, computation time, the physical cost of hardware, and its maintenance cost. As demonstrated in our experiments, a promising solution which minimizes the composite cost is sequential fusion, where a fusion algorithm sequentially uses match scores until a desired confidence is reached, or until all the match scores are exhausted, before outputting the final combined score., This evaluation was made possible thanks to the EU funding from the following projects: BioSecure (www.biosecure.info) and Mobio (www.mobioproject.org). The participating teams are supported by their respective national fund bodies: the Dutch BSIK/BRICKS project and the Spanish project TEC2006-13141-C03-03.

International audience; From state-of-the-art visualization algorithms, we distill six working principles which are, by hypothesis, sufficient to produce visual projections qualitatively similar to those obtained with these state-of-the-art algorithms. These working principles are presented through the geometrical reasoning of the classical Multidimensional Scaling algorithm, and their effectiveness is illustrated through a novel straightforward algorithm for data visualization. We show, using several datasets originated from various applications, that our algorithm can produce visual projections qualitatively similar to those obtained with these state-of-the-art algorithms. Besides, under the same motivation (of simplification), the problem of visualizing large datasets is tackled through a companion algorithm which is able to embed new input patterns.

Data rate and energy consumption are the major concerns in optical networks. In order to reduce energy consumption, transport operator networks based on optical circuit switching (OCS) concept, are becoming optically transparent, reducing optical to electrical (O/E) and electrical to optical (E/O) conversions. To face data rate increase, complex modulation formats and dual-polarization systems are considered and fiber spectrum is saved using network resources in a more efficient way, giving rise to a flexible frequency grid. Flexible transponders are developed to tune modulation formats and wavelengths and reconfigurable optical add/drop multiplexers (ROADMs) based on wavelength selective switches (WSSs) are studied. Flexible networks consider also a more dynamic traffic. Dynamism and flexibility lead to a deep transformation of the optical networks, including optical nodes, from both physical and control layer point of view. When channels are added and/or dropped, optical power excursion from erbium doped fiber amplifiers (EDFAs) has to be controlled dynamically. In that context, software defined networking (SDN) assisted by machine learning (ML) techniques is envisaged as promising candidate for the management and the dynamic control of optical networks. In this context, in the first part of our PhD work, we deal with optical power excursion in dynamic optical transport networks. In order to mitigate undesirable effects, we introduce and implement power excursion prediction and pre-compensation module using ML methods. As physical layer impairments (PLIs) accumulate along the path, we consider optical power excursion together with optical signal to noise ratio (OSNR) and bit error rate (BER), to estimate quality of transmission (QoT) of unseen channel configurations. Afterwards, using a reinforcement learning (RL) approach, we establish an autonomous impairment aware modulation format and wavelength assignment procedure, and we show that this permits to reduce the blocking probability of the incoming demands in optical nodes. In the second part of our PhD work, in the context of the N-GREEN project from the French national agency of research, we address a disruptive network architecture based on coloured optical packet switching (OPS). The main objective of N-GREEN is to propose a new generation of energy efficient routers. In the N-GREEN project, we perform the experimental characterization of an optical 2 x 2 switch based on semiconductor optical amplifiers (SOAs). This characterization leads to the proof of concept of a ring network with 10 nodes in cascade. Envisaging a 16 x 16 switch configuration, the experimental characterization, in single channel and WDM configurations, unveil interesting possibilities for the transmission of ultra-high data rates.; Le débit de données et la consommation d’énergie sont les principaux défis auxquels doivent faire face les réseaux optiques. Afin de réduire la consommation d’énergie, les réseaux opérateurs de transport optiques basés sur le concept de commutation de circuits optiques (OCS), deviennent optiquement transparents, réduisant les conversions optique/électrique (O/E) et électrique/optique (E/O). Pour faire face à l’augmentation du débit de données, on utilise des formats de modulations complexes et la technique de multiplexage en polarisation et on économise le spectre des fibres optiques en considérant une grille en longueurs d’onde plus flexible que la grille fixe ITU-T WDM. On développe des transpondeurs flexibles capables de sélectionner différents formats de modulation et longueurs d’onde; on développe également des multiplexeurs optiques d’insertion/extraction reconfigurables (ROADMs) basés sur des commutateurs sélectifs en longueur d’onde (WSSs). Ces réseaux flexibles prennent également en compte un trafic plus dynamique. Dynamisme et flexibilité impactent fortement les équipements des réseaux optiques, y compris les nœuds optiques d’un point de vue couche physique et couche de contrôle. Lorsque des canaux ou demandes sont ajoutés et/ou extraits, l’excursion de puissance optique des amplificateurs à fibre dopée à l’erbium (EDFAs) varie temporellement ce qui implique qu’elle doit être contrôlée dynamiquement. Dans ce contexte, le concept de réseau défini par le soft (SDN: Software Defined Network) prend tout son sens et l’introduction des techniques d’apprentissage machine (ML) permet d’entrevoir une aide au concept de SDN pour la gestion et le contrôle dynamique des réseaux optiques. Dans la première partie de ce travail de thèse, nous étudions l’excursion de puissance optique dans les réseaux de transport optiques dynamiques. Afin d’en atténuer les effets indésirables, nous introduisons et mettons en œuvre un module de prédiction et de pré-compensation de l’excursion de puissance en utilisant les méthodes ML. Comme les altérations de la couche physique (PLIs : Physical Layer Impairments) s’accumulent le long du chemin optique entre les noeuds source et destination de réseau, nous associons à l’excursion de puissance optique le rapport signal/bruit optique (OSNR: Optical Signal to Noise Ratio) et le taux d’erreur binaire (BER: Bit Error Rate), afin d’estimer la qualité de transmission (QoT: Quality of Transmission) de nouvelles configurations de canaux. Ensuite, en utilisant l’approche d’apprentissage par renforcement (RL), nous attribuons un format de modulation et une longueur d’onde aux différents canaux de façon automatique afin de réduire la probabilité de blocage des demandes entrantes dans les nœuds optiques. Dans la deuxième partie de ce travail de thèse, nous présentons notre contribution en tant que partenaire du projet ANR N-GREEN. Le principal objectif de N-GREEN est de proposer une nouvelle génération de routeurs peu consommateurs en énergie en considérant une architecture de réseau. Dans ce projet, nous abordons une architecture de réseau basée sur la commutation optique de paquets colorés (OPS: Optical Packet Switching) en rupture avec celle considérée dans la première partie de cette thèse. Dans le cadre de ce projet, nous avons caractérisé expérimentalement un commutateur optique 2 x 2 basé sur des amplificateurs optiques à semi-conducteurs (SOAs). Cette caractérisation nous a permis de valider un réseau en anneau constitué de 10 noeuds en cascade. En envisageant une configuration de commutateur 16 x 16, la caractérisation expérimentale, dans des configurations à canal unique et WDM, laisse entrevoir des possibilités intéressantes pour la transmission de données à très haut débit.

International audience

This study is part of the field of GNSS interference detection, in particular spoofing interferences. Spoofing consist in inducing a false position to a receiver, i.e a position different from the receiver's true position. The study's aim is to propose an approach for detecting spoofing interferences on an autonomous UAV using data directly available on the receivers. With the study of the state of the art of spoofing detection methods, the monitoring of the clock bias seemed like a potential method. A numerical modeling of a spoofing attack on a receiver showed that the clock bias undergoes jumps when it switch from the GNSS constellation to the spoofing constellation. By replicating this attack on a commercial receivers using real signals, the clock bias shows higher jumps than expected by the model and in some cases the clock drift also show some jumps. These jumps have been observed on different more or less subtle attack scenarios, however the amplitude of the jumps seems random.To go further than the simple spoofing detection, an approach using a communicating drone formation has been proposed. This method is based on a movement protocol allowing to delimit a space area where the spoofer is supposed to be located. The current protocol is not yet fully completed but it offers a promising basis.The study of the behavior of the clock bias highlighted its interest in a GNSS spoofing detection strategy. Based on this observation, further work could be carried out on the development and implementation on a UAV of a detection algorithm based on the monitoring of the clock bias. The study of the use of a drone formation for the localization of a spoofer led to the basis of a promising solution. Further work could be carried out in order to complete the movement protocol and to validate its efficiency against different types of spoofers.; La présente étude s'inscrit dans le domaine des interférences GNSS, en particulier les interférences de leurrage. Le leurrage consiste à induire une fausse position à un récepteur, c'est-à-dire une position différente de celle où il localisé. Cette étude consiste à proposer une approche de détection d'interférences de leurrage pour drone autonome utilisant les données directement issues des récepteurs. En réalisant un état de l'art des méthodes de détection de leurrage, le contrôle du biais d'horloge est apparu comme une approche potentielle. Une modélisation numérique d'une attaque de leurrage sur un récepteur a mis en évidence que le biais d'horloge présente des sauts lorsqu'il passe de la constellation GNSS à la constellation du leurre. En reproduisant cette attaque sur des récepteurs commerciaux utilisant de vrais signaux, le biais d'horloge présente des sauts plus importants que prévus par le modèle, et dans certains cas sur la dérive du biais également. Ces sauts ont été observés sur différents scénarios d'attaque plus ou moins subtiles, cependant l'amplitude de ces sauts semble aléatoire.Pour aller plus loin que la simple détection de leurrage, une approche utilisant une formation de drones communicants a été proposée dans le but de faire une estimation de la localisation du leurre. Cette méthode est basée sur un protocole de déplacement permettant à la formation de délimiter une zone de l'espace où le leurre est supposé être localisé. Le protocole actuel n'est pas encore complètement abouti mais il offre déjà une base prometteuse.L'étude du comportement du biais d'horloge a permis de mettre en évidence son intérêt dans une stratégie de détection de leurrage GNSS. A partir de ce constat, de futurs travaux pourront être menés sur le développement et l'implémentation sur un drone volant d'un algorithme de détection basé sur le contrôle du biais d'horloge. L'étude de l'utilisation d'une formation de drone pour la localisation d'un leurre a permis de poser les bases d'une solution prometteuse. De futurs travaux peuvent être menés afin de compléter le protocole de déplacement et de valider son efficacité face à différents types de leurres.

We propose a new method for the synthesis of arbitrary temporal output pulse profiles by using a birefringent laser pulse shaper. This shaper contains a series of N birefringent crystals placed between input and output polarizers. The synthesis method starts with the linear time-invariant system theory to determine the optimal amplitudes of the replica pulses at the output polarizer and the time delay between them, and then uses the temporal Jones matrix formalism to obtain the angle of each crystal. The design procedure is demonstrated for an eight-stage birefringent laser pulse shaper. Several examples of predefined output pulse profiles are given to show the potential of the proposed birefringent laser pulse shaper.

Identity or identification verification process using handwritten signatures affixed to a digital sensor Method for identifying or verifying the identity of a user, using a plurality of previously acquired reference signature vectors, a handwritten signature of the user and at least one additional handwritten information related to the user, previously affixed on a digital sensor, in particular a mobile sensor, method in which: a) said handwritten signature of the user and said at least one additional piece of information are merged to generate at least one test signature vector, b) said at least one signature vector test is compared to several of said reference signature vectors, and c) based at least on this comparison, a likelihood score is generated in order to identify or verify the identity of the user. Figure for the abstract: Fig. 1., Procédé d'identification ou de vérification de l'identité d'un utilisateur, utilisant une pluralité de vecteurs de signature de référence précédemment acquis, une signature manuscrite de l'utilisateur et au moins une information manuscrite complémentaire liée à l'utilisateur, préalablement apposées sur un capteur numérique, notamment mobile, procédé dans lequel : a) ladite signature manuscrite de l'utilisateur et ladite au moins une information complémentaire sont fusionnées pour générer au moins un vecteur de signature de test, b) ledit au moins un vecteur de signature de test est comparé à plusieurs desdits vecteurs de signature de référence, et c) sur la base au moins de cette comparaison, un score de vraisemblance est généré afin d'identifier ou de vérifier l'identité de l'utilisateur. Figure pour l'abrégé: Fig. 1.

A method for identifying or for verifying the identity of a user, using a plurality of previously acquired reference signature vectors, a handwritten signature of the user and at least one additional item of handwritten information linked to the user that are affixed beforehand to an in particular mobile digital sensor, in which method: a) said handwritten signature of the user and said at least one additional item of information are fused in order to generate at least one test signature vector, b) said at least one test signature vector is compared with a plurality of said reference signature vectors, and c) a likelihood score is generated on the basis at least of this comparison in order to identify or to verify the identity of the user., Procédé d'identification ou de vérification de l'identité d'un utilisateur, utilisant une pluralité de vecteurs de signature de référence précédemment acquis, une signature manuscrite de l'utilisateur et au moins une information manuscrite complémentaire liée à l'utilisateur, préalablement apposées sur un capteur numérique, notamment mobile, procédé dans lequel : a) ladite signature manuscrite de l'utilisateur et ladite au moins une information complémentaire sont fusionnées pour générer au moins un vecteur de signature de test, b) ledit au moins un vecteur de signature de test est comparé à plusieurs desdits vecteurs de signature de référence, et c) sur la base au moins de cette comparaison, un score de vraisemblance est généré afin d'identifier ou de vérifier l'identité de l'utilisateur.

International audience; This paper describes application of X-doped AlN (X=Sc, Cr and Y) to wideband surface acoustic wave (SAW) sensors in 200–300 MHz range. First, it is shown theoretically that Cr doped AlN thin film has the highest piezoelectric strain constant, accompanied by a lowest mechanical softening compared to Sc doped AlScN and Y doped AlN thin films for transition metals concentrations ranging from 0 to 25%. Next, the impact of transition metals (Sc, Cr and Y) concentration have been carried out for the first time, in terms of surface wave velocity, electrode reflectivity, transduction coefficient and distributed finger capacitance. Finely, the insertion loss of two-port SAW resonator based on AlXN (X=Sc, Cr and Y) deposited on sapphire substrate is obtained using P-matrix model, and it is shown that AlCrN-SAW resonator exhibit lower insertion loss compared to those based on AlScN and AlYN for metal concentrations of 25%. This finding may position Cr doped AlN as a prime piezoelectric material for low loss SAW sensors whose performance can be tuned via Cr composition.

Cyber-Physical Systems (CPS) use computational resources to control physical process and provide critical services. For this reason, an attack in these systems may have dangerous consequences in the physical world. Hence, resilience is a fundamental property to ensure the safety of the people, the environment and the controlled physical process. In this paper, we present metrics to quantify the resilience level based on the design, structure, stability, and performance under the attack of a given CPS. The metrics provide reference points to evaluate whether the system is better prepared or not to face the adversaries. This way, it is possible to quantify the ability to recover from an adversary using its mathematical model based on switched linear systems and actuators saturation. Finally, we validate our approach using a numeric simulation on the Tennesse Eastman control challenge problem., 11 pages, 2 figures, 6 tables

International audience; INTRODUCTION: One of the most significant challenges in the development of applications and AAL (Ambient Assisted Living) platforms is the interoperability between the various systems that compose the AAL platform. However, one solution would be to use the Internet of Things (IoT) technologies and standards to address requirements that, such as interoperability, are common to both domains. OBJECTIVES: To propose a model of sensor data representation and an IoT-based architecture in an AAL context, which reduces the impact of interoperability without overwhelming the overall system performance. METHODS: We have developed an AAL platform using pub/sub communication. A Raspberry Pi acts as a gateway among sensors, the RabbitMQ platform, and the AAL applications. We have developed a data representation model, based on existing AAL models, and a Protocol Bufferscommunication strategy that considers platform independence and spatial decoupling to address interoperability. RESULTS: We implemented and evaluated our model with the integration of two AAL applications installed in a Living Lab, which uses distinct devices, data formats, and programming languages. Then, we simulated a daily situation and analyzed the results regarding the overall performance. Results showed our solution presents a faster average response time than without its adoption. Also, our solution performed more stable behaviour.

International audience; Chip-scale frequency comb generators lend themselves as multi-wavelength light sources in highly scalable wavelength-division multiplexing (WDM) transmitters and coherent receivers. Among different options, quantum-dash (QD) mode-locked laser diodes (MLLD) stand out due to their compactness and simple operation along with the ability to provide a flat and broadband comb spectrum with dozens of equally spaced optical tones. However, the devices suffer from strong phase noise, which impairs transmission performance of coherent links, in particular when higher-order modulation formats are to be used. Here we exploit coherent feedback from an external cavity to drastically reduce the phase noise of QD-MLLD tones, thereby greatly improving the transmission performance. In our experiments, we demonstrate 32QAM WDM transmission on 60 carriers derived from a single QD-MLLD, leading to an aggregate line rate (net data rate) of 12 Tbit/s (11.215 Tbit/s) at a net spectral efficiency (SE) of 7.5 bit/s/Hz. To the best of our knowledge, this is the first time that a QD-MLLD optical frequency comb has been used to transmit an optical 32QAM signal. Based on our experimental findings, we perform simulations that show that feedback-stabilized QD-MLLD should also support 64QAM transmission with a performance close to the theoretical optimum across a wide range of technically relevant symbol rates.

International audience; Retrieving and indexing historical Arabic documents remain a very significant challenge. The purpose of this paper is to compare the feature representation spaces for word spotting in historical Arabic documents. Our goal is to create embedding spaces using the characteristics of different machine learning methods: i) linear such as principal component analysis and linear discriminant analysis, and ii) non-linear including convolutional neural networks for triplets and Siamese. Subsequently, each word image is represented by a dense vector. Thus, to match feature representations, a Euclidean distance is used. An evaluation of various representation space models is presented. The embedding word models are evaluated on the VML-HD dataset, and the experiments show the effectiveness of non-linear methods compared to linear ones.

International audience; BACKGROUND AND MOTIVATION: The world population is living longer than ever before, Ambient Assisted Living (AAL) research try to ensure people are ageing well by using digital and physical technologies. The development of AAL platforms presents several challenges (e.g., sensor data integration, device interoperability). Many requirements for AAL platforms can be met by using the Internet of Things (IoT) and Cloud Computing paradigm standards. METHODOLOGY: In this paper, we conducted a systematic mapping of the literature on AAL platforms to investigate the correlation between these three paradigms. We focused on AAL platforms that use IoT and Cloud Computing concepts to meet their requirements. Our primary research question is “What are the requirements of AAL platforms in the IoT Era and their relationship with IoT standards?”. We defined as the scope of our systematic mapping the primary studies about AAL platforms published between 2013 to 2017, and we used seven indexed electronic databases. RESULTS: We obtained 35 papers whose AAL platforms developed using IoT or Cloud Computing standards. We then analyse and correlate the requirements, architectures, and evaluation types presented by these platforms. We found patterns by connecting specific requirements with platform design. Webelieve these results assist AAL developers and architectural designers in the creation of new AAL platforms and applications.

International audience; This paper presents a new approach to the development of a radio frequency energy harvesting system. Starting from a sensor node to define the constraints of our system such as its energy requirement or its supply voltage, we develop a protocol. It allows us to validate whether it is possible to power such a system with an RF/DC converter. First we study a RF/DC converter capable of converting signals in the ISM band at 2.45 GHz whose ambient power has been previously characterized.

International audience; This paper presents a new approach to the development of a radio frequency energy harvesting system. Starting from a sensor node to define the constraints of our system such as its energy requirement or its supply voltage, we develop a protocol. It allows us to validate whether it is possible to power such a system with an RF/DC converter. First we study a RF/DC converter capable of converting signals in the ISM band at 2.45 GHz whose ambient power has been previously characterized.

International audience; TiO2/ZnO/TiO2/ZnO were grown on ordinary glass substrate by ultrasonic spray deposition at 300∘C. Titanium chloride, zinc acetate and deionized water were employed as precursors. The ultraviolet–visible spectrometer spectra show that the quality of optical transmittance in visible wavelength range decreases as the number of bilayers increases. The crystalline phase of the thin films was identified through X-ray diffraction method with diffraction patterns scan used at 2θ from 10∘ to 90∘. From the DRX results, it is found that the crystallinity is enhanced when the number of TiO2/ZnO layers increased. From the 2D/3D atomic force microscopy (AFM) analysis, it is found that the surface of the films was covered by triangle-like particles. However, bilayer and multilayer TiO2/ZnO thin films show smooth and compacted surface roughnesses compared to single-layer TiO2 and ZnO, respectively. In addition to this experimental work, a multichannel filter based on TiO2/ZnO subphotonic crystals is suggested. The transmission properties of this filter have been investigated by means of the transfer-matrix method (TMM). The idea is to import a wedge of air with an apex angle as the defect layer to disturb the original periodicity of the oxide photonic crystal and generate some new kind of multichannel filtering properties.

International audience; In this paper, we propose a P-Matrix model for the frequency analysis of SAW-delay line based on AlScN/Sapphire layered structures with 12.5 %, 25 % and 37.5 % scandium (Sc) concentrations. The effect of scandium concentration on the acoustic surface wave velocity, the electrode reflectivity, the transduction coefficient, the propagation loss and the distributed finger capacitance was taken into account in the frequency analysis of the layered SAW delay line. The acoustico-mechanical characterization indicated that the AlScN exhibit higher piezoelectric properties compared with those of pure AlN films, leading to significant improvement of electromechanical coupling coefficient. It also found that the Young’s modulus and the hardness of AlN film decrease significantly when Sc is doped, and is responsible for the decreased acoustic velocity. Additionally, the frequency analysis of AlScN-SAW delay line showed that the lower insertion loss predicted from the high Sc concentration, demonstrating the potential applications of AlScN in RF-MEMS applications. The resonant frequency also shifted slightly to the lower frequencies with increasing scandium concentration.

A novel reconfigurable dual-bandpass filter with independently controlled center frequencies and fixed bandwidths is presented. It was designed to be integrated as a radio frequency anti-aliasing filter in a reconfigurable multi-bandpass sampling global navigation satellite system (GNSS) receiver. The center frequency tuning ranges are 1190–1210 MHz and 1560–1590 MHz in the lower and the upper L-bands, respectively, with a fixed bandwidths of 35 MHz. Varactor tuning technology associated with open loop stub loaded resonator microstrip topology were investigated to achieve these goals. The proposed dual-bandpass filter design showed the agility to independently control the center frequencies in the two L-bands and to deactivate one of the passbands. Furthermore, it shows enough flexibility to adapt to the deployment of new GNSS standards specifications. It exhibits on all the tuning ranges an insertion and a return losses of less than 4.5 dB and more than 10 dB, respectively.

International audience

Large amounts of ever-increasing global data traffic require sound and reliable communication channels. Optical terrestrial networks and submarine links are at the very heart of the global telecom infrastructure, and carry hundreds of frequency channels modulated at very high rates. These links include not only kilometers of fiber but also optical elements such as Erbium-Doped Fiber-Amplifiers (EDFA) to amplify periodically the attenuated signals and Wavelength Selective Switches (WSS) to route the signals to their assigned destinations. In this thesis, we explore a specific rate-degrading impairment of the optical propagation that raises in those systems. Discrete optical elements often exhibit a polarization anisotropy that emerges as a gain or loss imbalance between the two polarization tributaries of the polarization-multiplexed optical signals. This non-unitary effect called Polarization Dependent Loss (PDL) impairs the quality of transmission in current and next-generation optical systems.In the context of polarization-multiplexed signals, we assess the capacity loss induced by PDL. First, channel models are carefully studied and two approaches are described: a channel with distributed PDL elements and distributed noise or a simplified single-element equivalent channel. Making use of these models, we then analyze their fundamental limits of communications. We show that the PDL channel capacity depends in practice on the state-of-polarization orientation of the incident signal. We then review the state-of-the art of PDL-mitigating modulation schemes and propose two new multi-dimensional signaling schemes that enhance worst-case and average performance. These two modulations are unitary transforms of M-QAM symbols and do not make use of additional degrees of freedom apart from the four already-used dimensions per wavelength (in-phase and quadrature channels of two polarization states). We briefly extend these first results to space-division-multiplexed optical communications impaired by mode dependent loss (MDL) that present a similar gain imbalance. Beyond signal shaping at the transmitter side, we study the performance loss of a conventional, sequential signal processing chain at the receiver side in presence of PDL, in comparison with a joint equalization-decoding scheme. The additional capacity loss due to the mismatch sequential processing is evaluated for several modulation formats and at different operating points. Finally, we report an experimental validation of the two proposed signaling schemes, both on a single PDL element and on a distributed PDL channel, demonstrating the predicted enhanced robustness to PDL., La demande de débit en augmentation constante requiert des canaux de communication fiables et robustes. Les réseaux optiques terrestres ainsi que les liens sous-marins représentent le cœur de l'infrastructure des télécommunications au niveau mondial, et transportent des centaines de canaux fréquentiels modulés à haut débit. Ces liens se composent non seulement de kilomètres de fibre mais aussi de composants optiques comme les amplificateurs à fibre dopée à l'erbium (EDFA) pour relever régulièrement la puissance des signaux atténués, ou encore des commutateurs sélectifs en longueur d'onde (WSS) qui routent le signal vers leur destination. Dans cette thèse, nous nous intéressons à une pénalité spécifique qui survient dans ces systèmes et qui réduit le débit d'information dans une propagation optique. Les composants optiques discrets présentent typiquement une anisotropie en polarisation : un déséquilibre de perte (ou gain) apparaît entre les deux polarisations d'un signal optique multiplexé en polarisation. Cet effet non unitaire appelé pertes dépendant de la polarisation (abrégé PDL en anglais) dégrade la qualité de transmission dans les systèmes optiques actuels et futurs.Dans le cadre des transmissions multiplexés en polarisation, nous caractérisons la perte de capacité induite par la PDL. Pour commencer, des modèles de canal sont attentivement étudiés et deux approches sont présentées : un canal avec des éléments PDL distribués avec un bruit également distribué, ou alors un canal équivalent simplifié en un seul élément. Nous analysons les limites fondamentales de communication associées à ces modèles. Nous montrons en outre que la capacité d'un canal PDL dépend en pratique de l'orientation de l'état de polarisation du signal incident. Nous passons ensuite en revue l'état de l'art de schémas de modulation atténuant l'effet de PDL puis nous proposons deux nouveaux schémas de modulation multi-dimensionnels qui augmentent la performance minimum et moyenne du canal. Ces deux modulations sont des transformations unitaires de symboles M-QAM qui n'utilisent pas de degrés de liberté autres que les quatre déjà utilisés pour chaque longueur d'onde (encodage en phase et quadrature de phase sur les deux états de polarisation). Nous proposons succinctement une extension de ces premiers résultats à des communications optiques plus généralement multiplexés en dimension spatiale et impactées par des pertes dépendantes de mode, qui présentent un déséquilibre de gain identique. Au-delà de la construction de modulations au niveau du transmetteur, nous étudions la perte de performance en présence de PDL d'une chaîne de traitement de signal conventionnelle et séquentielle par comparaison avec une égalisation et décodage joint. Cette perte additionnelle de capacité résultant du traitement séquentiel non optimal est évaluée pour différents formats de modulations ainsi que différents points de fonctionnement. Enfin, nous exposons une validation expérimentale des deux schémas de modulation proposés, aussi bien sur un élément de PDL seul ou sur un canal PDL distribué, et montrons comme attendu une robustesse augmentée vis-à-vis de la PDL.

International audience; Hashing has been widely applied to approximate nearest neighbor search for large-scale multimedia retrieval. A variety of hashing methods have been developed for learning an efficient binary data representation, mainly by relaxing some imposed constraints during hash function learning. Although they have achieved good accuracy-speed trade-off, the resulting binary codes may fail sometimes in adequately approximating the input data, thus significantly decreasing the search accuracy. In this paper, we present a new Unsupervised Deep Learning Hashing approach, called Deep Neuron-per-Neuron Hashing, for high dimensional data indexing. Unlike most existing hashing approaches, our method does not seek to binarize the neural network output, but rather relies directly on the continuous output to create an efficient index structure with hash tables. Given the neural network deepest layer, each table indexes separately a neuron output, capturing in this way a particular high level individual structure (feature) of the input. An efficient search is then performed by computing a cumulative collision score of a given query over all the neuron-based hash tables. Experimental comparisons to the state-of-the-art demonstrate the competitiveness of the proposed method for large datasets

International audience; Optical code division multiple access (OCDMA) systems are attracting an increasing interest in optical fiber communication. This is due to the various advantages that they provide; in particular for the simultaneous improvement capacity and flexible channel allocation as well as fastest data speeds and increased security. In this paper, the authors present an experimental demonstration of bipolar optical CDMA System with phase shift using E-beam technique on Hx Li1−x NbO3 transmission channel realized by proton exchange. The idea of the proposed system is to evaluate the result of eight cascaded Bragg filters in very narrow band. The coded sequence corresponds to that of Hadamard H8(7), represented by the bits ’1 1 −1 −1 −1 −1 1 1’, and phase shift have been identified on the coder spectral response whilst increasing the multiplexing capacity in terms of the total number of users and the data.

International audience; All sorts of positioning system exist and there is a growing demand nowadays in order to provide users with additional data in many activities. Horce racing is no exception and the PMU (Paris Mutuel Urbain) is in charge of driving the project concerning the deployment of a tracking system. PMU is Europe's largest betting operator. It is an Economic Interest Grouping (EIG), whose mission is to finance the French horce racing industry. The specifications of the tracking system to be implemented are quite tough : an accuracy of positioning of 25 cm for more than 98 % of the time during races for all the horses. A call for proposals has been issued and a few competitors have been hired in order to demonstrate the real performances on their system in a so-called "pilot" phase. The problem we now have is how to evaluate the real accuracy of the system in real conditions,i.e., during "simulated races" Additional aspects are also of uppermost importance, such as the latency and the way data are displayed, but the present paper will only focus on the evaluation methodology used for the positioning accuracy assessment. An incremental validation approach was set up in order to allow the competitors to gradually improve their solutions, from the "easiest" tests (with cars) to the "most difficult" ones on horses during simutated races. Note that all competitors proposed a GNSS (Global Navigation Satellite System)-based solution using an RTK (Real-Time Kinematic) approach

International audience; This study focused on two plant species endemic to the Algerian Sahara: Artemisia campestris ssp. glutinosa (J. Gay) Batt. and Artemisia judaïca ssp. sahariensis (Chev.). The aim of this work was to determine their chemical composition for a better insight into their traditional use and to evaluate the antibacterial potential of the essential oils of their leafy stems on 8 different bacterial strains. The analysis revealed that these two Artemisia species contain, respectively, 14.5 ± 0.40 % and 11.4 ± 0.45 % of total nitrogenous matter, 17.7 ± 0.34 % and 33.7 ± 0.49 % of crude fiber, 9.5 ± 0.82 % and 8.5 ± 0.79 % of mineral matter. Moreover, the results showed that the essential oils of Artemisia campestris ssp. Glutinosa contained α-pinene (8.83 ± 1.02 %), β-pinene (28.94 ± 1.16 %), limonene (8.51 ± 0.19 %), α-thujone (0.18 ± 0.06 %), β-thujone (0.67 ± 0.15 %) and camphor (0.86 ± 0.02 %). These essential oils also showed a high inhibitory activity on the growth of Enterococcus faecium, Staphylococcus pneumoniae and Staphylococcus aureus S. On the other hand, the essential oils of Artemisia judaïca ssp. sahariensis mainly contained piperitone (65.88 ± 6.12%), davanone (12.58 ± 5.81 %) and thymol (0.56 ± 0.06 %) and they exhibited an inhibitory activity on Pseudomonas aeruginosa S, Proteus mirabilis and Klebseilla pneumoniae. These results demonstrated the inhibitory activity of Artemisia's essential oils against certain human pathogens and thus they showed the great interest of studying Artemisia species for the scientific and medical community

International audience; This paper presents the design and a first evaluation of a new monitoring system based on contactless sensors to estimate sleep quality. This sensor produces thermal signals which have been used, at first, to detect a human presence in the bed and then to estimate sleep quality. To distinguish between different sleep phases, we have used methods of signal processing in order to extract the necessary features for learning an adapted statistical model. The existing monitoring systems use sensors attached to the bed or worn by the person. We propose in this paper a system based on a passive thermal sensor which has the advantage of being fixed on the wall, thus it is easier to use and more reliable. We explain different signal processing steps and describe sleep stage recognition algorithms. We propose an adaptation of the SAX method for the thermal signal. Finally, we evaluate our system in comparison with a polysomnographic recording system in the Hospital (CHU) of Limoges.

We report the optical and electronic characteristics of GaP1−xBix bismide alloys with dilute composition 0 ≤ x ≤ 0.187 in the cubic zinc blende crystallographic phase calculated using a scalar relativistic Full Potential Linear Augmented Plane Wave method within Density Functional Theory. The equilibrium lattice constants are calculated using the improved generalized gradient approximation of Wu and Cohen in excellent agreement with experimental values. A quasi-linear dependence of the lattice parameters on Bi composition is obtained. Electronic structures such as the band gap, total density of states and partial density of states have been discussed using the recently developed approximate modified Becke–Johnson of Tran-Blaha with spin–orbit interaction (SOI) included. Electronic band structure analysis shows a significant decrease of band gap energy when P is replaced by Bi (by ∼548 meV per x = 0.062 ). The band gap can be tuned in the emission wavelengths range from the visible to near-infra spectrum (∼0.414–1.633 μm for x ∼ 0.187 ). The decrease in band gap energy is attributed to the both resonance interactions between Bi_p sates with the top of the valence band and Ga_s states at the host conduction band when the Bi content increased. The band gap undergoes an indirect (Γ–X) to direct (Γ–Γ) transition at x = 0.05 . The complex dielectric function, refractive index, reflectivity, extinction and absorption coefficient were calculated for incident radiation energy 0–8 eV. The critical-points in various optical spectra are identified in excellent agreement with measured values. The maximum absorption occurs within the energy range from ∼5 to 8 eV, corresponding to the ultraviolet spectrum region. Dilute GaP1−xBix alloys are a promising material system for the realization of the future photonic devices operating in the near-infra spectrum.

Detecting the early signs of diabetic retinopathy (DR) is essential, as timely treatment might reduce or even prevent vision loss. Moreover, automatically localizing the regions of the retinal image that might contain lesions can favorably assist specialists in the task of detection. In this study, we designed a lesion localization model using a deep network patch-based approach. Our goal was to reduce the complexity of the model while improving its performance. For this purpose, we designed an efficient procedure (including two convolutional neural network models) for selecting the training patches, such that the challenging examples would be given special attention during the training process. Using the labeling of the region, a DR decision can be given to the initial image, without the need for special training. The model is trained on the Standard Diabetic Retinopathy Database, Calibration Level 1 (DIARETDB1) database and is tested on several databases (including Messidor) without any further adaptation. It reaches an area under the receiver operating characteristic curve of 0.912-95%CI(0.897-0.928) for DR screening, and a sensitivity of 0.940-95%CI(0.921-0.959). These values are competitive with other state-of-the-art approaches.

The invention relates to a method for generating at least one information about the production of a handwritten, hand-affixed or printed trace on a surface, comprising: - extracting several features describing the trace from at least one three- dimensional image of said trace, acquired by an imagery system, and - inputting said extracted features in a trained module to output said at least one information, said module having been trained beforehand with a plurality of previously- acquired three-dimensional images of traces and corresponding information related to the production of these traces., L'invention concerne un procédé de génération d'au moins une information concernant la production d'un tracé manuscrit, apposé manuellement ou imprimé sur un surface, comportant les étapes consistant à: - extraire plusieurs caractéristiques décrivant le tracé à partir d'au moins une image tridimensionnelle dudit tracé, acquise par un système d'imagerie, et - introduire lesdites caractéristiques extraites dans un module entraîné pour délivrer ladite ou lesdites informations, ledit module ayant été entraîné préalablement avec une pluralité d'images tridimensionnelles acquises précédemment de tracés et des informations correspondantes liées à la production de ces tracés.