Publications

2018

• Speed Index: Relating the Industrial Standard for User Perceived Web Performance to web QoE
  
  • Hossfeld Tobias
  • Metzger Florian
  • Rossi Dario
  , 2018.
• Broadband High Gain Polarization-Maintaining Holmium-doped Fiber Amplifiers
  
  • Tench Robert
  • Romano Clément
  • Delavaux Jean-Marc
  , 2018, pp.paper Mo3E.3.
• Dual Polarization nonlinear frequency division multiplexing transmission
  
  • Gemechu Wasyhun A.
  • Gui Tao
  • Song Mengdi
  • Wabnitz Stephano
  • Hafermann Hartmut
  • Tao Lau Alan Pak
  • Yousefi Mansoor
  • Jaouën Yves
  IEEE Photonics Technology Letters, Institute of Electrical and Electronics Engineers, 2018, 30 (18), pp.1589-1592.
• Optimal Deployment of Security Policies: Application to Industrial Control Systems
  
  • Ismail Ziad
  • Leneutre Jean
  • Fourati Alia
  , 2018.
• On Binary Classification in Extreme Regions
  
  • Jalalzai Hamid
  • Clémençon Stéphan
  • Sabourin Anne
  , 2018. In pattern recognition, a random label Y is to be predicted based upon observing a random vector X valued in R d with d ≥ 1 by means of a classification rule with minimum probability of error. In a wide variety of applications, ranging from finance/insurance to environmental sciences through teletraffic data analysis for instance, extreme (i.e. very large) observations X are of crucial importance, while contributing in a negligible manner to the (empirical) error however, simply because of their rarity. As a consequence, empirical risk minimizers generally perform very poorly in extreme regions. It is the purpose of this paper to develop a general framework for classification in the extremes. Precisely, under non-parametric heavy-tail assumptions for the class distributions, we prove that a natural and asymptotic notion of risk, accounting for predictive performance in extreme regions of the input space, can be defined and show that minimizers of an empirical version of a non-asymptotic approximant of this dedicated risk, based on a fraction of the largest observations, lead to classification rules with good generalization capacity, by means of maximal deviation inequalities in low probability regions. Beyond theoretical results, numerical experiments are presented in order to illustrate the relevance of the approach developed
• The Forward-Backward Embedding of Directed Graphs
  
  • de Lara Nathan
  • Bonald Thomas
  openreview.net, 2018.
• Compressed Sensing for Wideband HF Channel Estimation
  
  • Crespo Marques Elaine
  • Maciel Nilson
  • Naviner Lirida
  • Cai Hao
  • Yang Jun
  , 2018, pp.1-5.
• A Popularity-Based Approach for Effective Cloud Offload in Fog Deployments
  
  • Enguehard Marcel
  • Carofiglio Giovanna
  • Rossi Dario
  , 2018.
• On the Performance and Security of Multiplication in GF(2N)
  
  • Danger Jean-Luc
  • El Housni Youssef
  • Facon Adrien
  • Gueye Cheikh T.
  • Guilley Sylvain
  • Herbel Sylvie
  • Ndiaye Ousmane
  • Persichetti Edoardo
  • Schaub Alexander
  Cryptography, 2018. (10.3390/cryptography2030025)
  DOI : 10.3390/cryptography2030025
• 8W 1952nm Highly Efficient Brillouin-free Single Clad TDFA
  
  • Romano Clément
  • Tench Robert
  • Delavaux Jean-Marc
  , 2018, pp.paper We2.5.
• Statistical properties of side-channel and fault injection attacks using coding theory
  
  • Carlet Claude
  • Guilley Sylvain
  Cryptography and Communications - Discrete Structures, Boolean Functions and Sequences, Springer, 2018, 10 (5), pp.909-933. Naïve implementation of block ciphers are subject to side-channel and fault injection attacks. To deceive side-channel attacks and to detect fault injection attacks, the designer inserts specially crafted error correcting codes in the implementation. The impact of codes on protection against fault injection attacks is well studied: the number of detected faults relates to their minimum distance. However, regarding side-channel attacks, the link between codes and protection efficiency is blurred. In this paper, we relate statistical properties of code-based countermeasures against side-channel attacks to their efficiency in terms of security, against uni- and multi-variate attacks. (10.1007/s12095-017-0271-4)
  DOI : 10.1007/s12095-017-0271-4
• Snaky Leaky Wave Antenna for Scanning Applications in W Band
  
  • Bernabeu-Jimenez Tomas
  • Begaud Xavier
  • Magne François
  • Hadjloum Massinissa
  • Cosson Bruno
  , 2018.
• A Fast and Scalable Fragmentation Algorithm For Data Protection For Multi-Storage Over Independent Locations
  
  • Kapusta Katarzyna
  • Memmi Gérard
  , 2018.
• High Optical Budget 25Gbit/s PON with PAM4 and Optically Amplified O Band Downstream Transmission
  
  • Barthomeuf Sylvain
  • Saliou Fabienne
  • Anet Neto Luiz
  • Chanclou Philippe
  • Erasme Didier
  , 2018 (Mo4B.1). We experimentally demonstrate real-time downstream PAM4 transmission in O-band over 20km achieving a 35.2dB Optical Budget. Using a PDFA at the OLT and a SOA at the ONU increases the OB and eases the threshold decision of the PAM4 decoder.
• An Adaptive Primal-Dual Framework for Nonsmooth Convex Minimization
  
  • Tran-Dinh Quoc
  • Alacaoglu Ahmet
  • Fercoq Olivier
  • Cevher Volkan
  Mathematical Programming Computation, 2018.
• Neurophysiological Profile of Antismoking Campaigns
  
  • Modica Enrica
  • Cartocci Giulia
  • Perrotta Davide
  • Di Feo Paolo
  • Mancini Marco
  • Aricò Pietro
  • M. S. Inguscio Bianca
  • Babiloni Fabio
  • Rossi Dario
  Computational Intelligence and Neuroscience, 2018.
• Single Receiver Multi-antenna Code-Modulated IR-UWB Front-End
  
  • Vo Tien Tu
  • Ouvry Laurent
  • Sibille Alain
  , 2018. The multi-antenna receiver has many use cases in wireless communications and Radar but it faces some major drawbacks such as high cost of hardware and high power consumption. Recent developments on narrow band systems have led to an alternative solution called Code-Division Multiplexing (CDM), using a single RF receiver front-end to receive signals from multi-antenna. For this purpose signals from the antennas are mixed with high speed spreading codes and then are combined at the RF stage. However, the use of high speed codes increases the signal bandwidth. Applying this solution on Ultra-Wide Band (UWB) systems is questionable because of the bandwidth increase. This paper discusses the principles and implementation of CDM technique on an Impulse Radio (IR) UWB system showing interesting insights. An experiment is performed to estimate the Angle-Of-Arrival (AOA) on the recovered signals in a basic indoor positioning scenario. (10.1109/PIMRC.2018.8580916)
  DOI : 10.1109/PIMRC.2018.8580916
• Teoria da informação e da codificação
  
  • Rioul Olivier
  , 2018, pp.288 pp.. Livro de referência que trata da teoria da informação em detalhes, desde a apresentação das ferramentas básicas da teoria (entropia, divergência, informação mútua, teorema de tratamento de dados, informação de Fisher e variância entrópica) até a demonstração dos teoremas de Shannon (para a codificação de fonte com ou sem perdas, a codificação de canal e a codificação conjunta fonte/canal). Seu público-alvo são os pesquisadores e engenheiros de telecomunicações, assim como os estudantes universitários interessados no tema.
• Dynamic and Noise Properties of High-Q Hybrid Laser
  
  • Gallet Antonin
  • Grillot Frédéric
  , 2018.
• Statistical methods for joint antenna-radio channel modelling
  
  • Sibille Alain
  , 2018. The description of the electromagnetic behavior of antennas needs a large amount of data to be complete, since it requires complex numbers for the radiated far field in all possible directions and polarizations, for all frequencies of interest and all antenna elements in case of arrays or multiport antennas, plus the full impedance matrix. In addition, since most of the time the radio channel linking the transmitting and receiving antennas in a wireless communication is not in free space, many propagation effects are involved and result in the particularities of the received signals. Finally, while it is often designed and measured in nearly ideal conditions (e.g. anechoic chamber) an antenna is rarely used in emptiness and its close environment does impact its performance, sometimes dramatically. While strong close disturbers can be taken into account at the design phase (such as a casing or a human head), variations in the effective impact of such disturbers can take place, which to some extent can unpredictably affect the antenna characteristics. Deterministic methods can hardly take into account all the variabilities that occur in real life and would be much too expensive to implement and to use. The natural approach, widely practiced in other domains, is to resort to statistical descriptions, based on metamodels able to represent these variabilities with a limited number of parameters while achieving an adequate trade-off between accuracy and simplicity. The presentation will address these issues for joint antenna-channel modelling, giving some examples about the development and use of such methods.
• Somatic Mutations Render Human and Pathogen Exomes more Similar
  
  • Engler Maggie
  • Tchamkerten Aslan
  • Cristescu Razvan
  PLoS ONE, Public Library of Science, 2018.
• Entrainment phenomenon in a mid-infrared QCL with external optical feedback and low frequency modulation
  
  • Spitz O
  • Wu J
  • Carras M
  • Wong C W
  • Grillot F
  , 2018. This experimental work reports on mid-infrared quantum cascade lasers operating under external optical feedback and external periodic bias forcing. A way of controlling spikes occuring in a deterministic chaotic pattern is experimentally described. These findings are of paramount importance for future optical countermeasure systems and secure atmospheric transmission lines as well as for reproducing neuronal systems and the communication between neurons due to sudden bursts.
• ANALYSIS OF COMMON DESIGN CHOICES IN DEEP LEARNING SYSTEMS FOR DOWNBEAT TRACKING
  
  • Fuentes Magdalena
  • Mcfee Brian
  • Crayencour Hélène C
  • Essid Slim
  • Bello Juan P
  , 2018. Downbeat tracking consists of annotating a piece of musical audio with the estimated position of the first beat of each bar. In recent years, increasing attention has been paid to applying deep learning models to this task, and various architectures have been proposed, leading to a significant improvement in accuracy. However, there are few insights about the role of the various design choices and the delicate interactions between them. In this paper we offer a systematic investigation of the impact of largely adopted variants. We study the effects of the temporal granularity of the input representation (i.e. beat-level vs tatum-level) and the encoding of the networks outputs. We also investigate the potential of convolutional-recurrent networks, which have not been explored in previous downbeat tracking systems. To this end, we exploit a state-of-the-art recurrent neural network where we introduce those variants, while keeping the training data, network learning parameters and post-processing stages fixed. We find that temporal granularity has a significant impact on performance, and we analyze its interaction with the encoding of the networks outputs.
• Event Detection in Wireless Body Area Networks Using Kalman Filter and Power Divergence
  
  • Salem Osman
  • Serhrouchni Ahmed
  • Mehaoua Ahmed
  • Boutaba Raouf
  IEEE Transactions on Network and Service Management, IEEE, 2018, 15 (3), pp.1018-1034. The collected data by biomedical sensors must be analyzed for automatic detection of physiological changes. The early identification of an event in collected data is required to trigger an alarm upon detection of patient health degradation. Such alarms inform healthcare professionals and allow them to quickly react by taking appropriate actions. However, events result from physiological change or faulty measurements, and lead to false alarms and unnecessary medical intervention. In this paper, we propose a framework for automatic detection of events from collected data by biomedical sensors. The proposed approach is based on the Kalman filter to forecast the current measurement and to derive the baseline of the time series. The power divergence is used to measure the distance between the forecasted and measured values. When a change occurs, this metric significantly deviates from past values. To distinguish emergency events from faulty measurements, we exploit the spatial correlation between the monitored attributes. We conduct experiments on real physiological data set and our results show that our proposed framework achieves a good detection accuracy with a low false alarm rate. Its simplicity and processing speed make our proposed framework efficient and effective for real-world deployment. (10.1109/TNSM.2018.2842195)
  DOI : 10.1109/TNSM.2018.2842195
• Design, Fabrication and Characterization of Hybrid III-V/SOI Phase-shift Free DFB Laser with Tapered Silicon waveguide
  
  • Gallet Antonin
  • Grillot Frédéric
  • Al. Et
  , 2018.