Publications

2018

• Low frequency fluctuations and optical bursts induced by external optical feedback and low frequency modulation in a quantum cascade laser at 9 µm
  
  • Spitz O
  • Wu J
  • Carras M
  • Wong C W
  • Grillot F
  , 2018. This experimental work reports on the first observation of low frequency fluctuations in a quantum cascade laser emitting at 9 µm, in one of the transparency windows of the atmosphere, 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 prime 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.
• A Dynamic Scalable Blockchain Based Communication Architecture for IoT
  
  • Qiu Han
  • Qiu Meikang
  • Liu Meiqin
  • Ming Zhong
  • Memmi Gérard
  , 2018, LNCS 11371, pp.159-166.
• Statistical inference for the Russell measure of technical efficiency
  
  • Badunenko Oleg
  • Mozharovskyi Pavlo
  Journal of the Operational Research Society, 2018.
• Signal and Noise Detection using Recurrent Autoencoders on Seismic Marine Data
  
  • Chambefort Mathieu
  • Chautru Emilie
  • Clémençon Stéphan
  • Poulain Guillaume
  • Salaun Nicolas
  , 2018.
• Random forests for resource allocation in 5G cloud radio access networks based on position information
  
  • Imtiaz Sahar
  • Koudouridis Georgios
  • Ghauch Hadi
  • Gross James
  EURASIP Journal on Wireless Communications and Networking, SpringerOpen, 2018, 2018 (1). (10.1186/s13638-018-1149-7)
  DOI : 10.1186/s13638-018-1149-7
• Building the Case for Temperature Awareness in Energy Consumption Models: an Application of the Energy-Frequency Convexity Rule
  
  • Vaddina Kameswar
  • Brandner Florian
  • Memmi Gerard
  • Jouvelot Pierre
  , 2018. Optimizing computing and communication systems that host energy-critical applications is becoming a key issue for software developers. In previous work, we introduced and validated the Energy/Frequency Convexity Rule for CPU-bound benchmarks on recent ARM platforms. This rule states that there exists an optimal clock frequency that minimizes the CPU's energy consumption for non-performance-critical programs. We showed that the Energy/Frequency Convexity Rule is related to the non-linearity of power with respect to frequency and is not dependent on the supply voltage. Here, we discuss the application of an analytical energy consumption model proposed previously to our target board, a TI AM572x EVM. We show that this non-linear analytical model can, for our experimental settings, be approximated by a frequency-linear variant, as our voltage is maintained constant. This, however, does not fit the measurements on the board, suggesting that a parameter is currently missing in the analytical model. We conjecture that accounting for temperature in the model would yield more accurate results that are in-line with our measurements. This builds the case for the inclusion of this important parameter in our energy models.
• Statistical electromagnetics for antennas
  
  • Acikgoz Hulusi
  • Arya Ravi Kumar
  • Wiart Joe J
  • Mittra Raj
  , 2018, pp.259-286. (10.1049/SBEW543G_ch8)
  DOI : 10.1049/SBEW543G_ch8
• Development and implementation of compressed sensing-based denoising and acquisition strategies for fluorescence microscopy and optical coherence tomography
  
  • Meiniel William
  , 2018. The mathematical theory of Compressed Sensing (CS) is a recently developed framework that enables the reconstruction of a signal or an image from very few measurements. In this thesis, we investigate how this theory can be implemented in the context of two optical microscopy techniques : fluorescence microscopy, and optical coherence tomography. Both technologies present different limitations which we prove can be tackled by the embedding of CS driven strategies. The latter can be divided into two categories : image processing algorithmic solutions, and optical acquisition techniques.
• Nonlinear models for neurophysiological time series
  
  • Dupré La Tour Tom
  , 2018. In neurophysiological time series, strong neural oscillations are observed in the mammalian brain, and the natural processing tools are thus centered on narrow-band linear filtering.As this approach is too reductive, we propose new methods to represent these signals.We first focus on the study of phase-amplitude coupling (PAC), which consists in an amplitude modulation of a high frequency band, time-locked with a specific phase of a slow neural oscillation.We propose to use driven autoregressive models (DAR), to capture PAC in a probabilistic model. Giving a proper model to the signal enables model selection by using the likelihood of the model, which constitutes a major improvement in PAC estimation.%We first present different parametrization of DAR models, with fast inference algorithms and stability discussions.Then, we present how to use DAR models for PAC analysis, demonstrating the advantage of the model-based approach on three empirical datasets.Then, we explore different extensions to DAR models, estimating the driving signal from the data, PAC in multivariate signals, or spectro-temporal receptive fields.Finally, we also propose to adapt convolutional sparse coding (CSC) models for neurophysiological time-series, extending them to heavy-tail noise distribution and multivariate decompositions. We develop efficient inference algorithms for each formulation, and show that we obtain rich unsupervised signal representations.
• Multimode Photonic Spectral Encoding for Single Qubit Arbitrary Unitary Synthesis
  
  • Raghunathan Ravi
  • Guillaume Ricard
  • Miatto Filippo M
  • Zaquine Isabelle
  • Alleaume Romain
  , 2018.
• Random monotone operators and application to stochastic optimization
  
  • Salim Adil
  , 2018. This thesis mainly studies optimization algorithms. Programming problems arising in signal processing and machine learning are composite in many cases, i.e they exhibit constraints and non smooth regularization terms. Proximal methods are known to be efficient to solve such problems. However, in modern applications of data sciences, functions to be minimized are often represented as statistical expectations, whose evaluation is intractable. This cover the case of online learning, big data problems and distributed computation problems. To solve this problems, we study in this thesis proximal stochastic methods, that generalize proximal algorithms to the case of cost functions written as expectations. Stochastic proximal methods are first studied with a constant step size, using stochastic approximation techniques. More precisely, the Ordinary Differential Equation method is adapted to the case of differential inclusions. In order to study the asymptotic behavior of the algorithms, the stability of the sequences of iterates (seen as Markov chains) is studied. Then, generalizations of the stochastic proximal gradient algorithm with decreasing step sizes are designed to solve composite problems. Every quantities used to define the optimization problem are written as expectations. This include a primal dual algorithm to solve regularized and linearly constrained problems and an optimization over large graphs algorithm.
• Storage, computation, communication: a fundamental tradeoff in distributed computing
  
  • Yan Qifa
  • Yang Sheng
  • Wigger Michèle
  , 2018. We consider a MapReduce-like distributed computing system. We derive a lower bound on the communication cost for any given storage and computation costs. This lower bound matches the achievable bound we proposed recently. As a result, we completely characterize the optimal tradeoff between the storage, the computation, and the communication. Our result generalizes the previous one by Li et al. to also account for the number of computed intermediate values. (10.1109/itw.2018.8613519)
  DOI : 10.1109/itw.2018.8613519
• On the Capacity of MIMO Optical Wireless Channels
  
  • Li Longguang
  • Moser Stefan M
  • Wang Ligong
  • Wigger Michèle
  , 2018, pp.1-5. This paper investigates the capacity of the multiple-input multiple-output free-space optical intensity channel under a per-input-antenna peak-power constraint and a total average-power constraint over all input antennas. Our work considers the setup with more transmit than receive antennas, and characterizes capacity as an alternative optimization problem over the distribution of the input vector times the channel matrix. This alternative capacity expression is then used to obtain upper and lower bounds on the capacity, which match asymptotically in the high signal-to-noise ratio regime. (10.1109/itw.2018.8613496)
  DOI : 10.1109/itw.2018.8613496
• Diffusion approximations and control variates for MCMC
  
  • Brosse Nicolas
  • Durmus Alain
  • Meyn Sean
  • Moulines Éric
  , 2018. A new methodology is presented for the construction of control variates to reduce the variance of additive functionals of Markov Chain Monte Carlo (MCMC) samplers. Our control variates are defined as linear combinations of functions whose coefficients are obtained by minimizing a proxy for the asymptotic variance. The construction is theoretically justified by two new results. We first show that the asymptotic variances of some well-known MCMC algorithms, including the Random Walk Metropolis and the (Metropolis) Unadjusted/Adjusted Langevin Algorithm, are close to the asymptotic variance of the Langevin diffusion. Second, we provide an explicit representation of the optimal coefficients minimizing the asymptotic variance of the Langevin diffusion. Several examples of Bayesian inference problems demonstrate that the corresponding reduction in the variance is significant, and that in some cases it can be dramatic.
• Mixed Delay Constraints at Maximum Sum-Multiplexing Gain
  
  • Nikbakht Homa
  • Wigger Michèle
  • Shamai Shitz Shlomo
  , 2018, pp.1-5. Coding schemes are proposed for Wyner's soft-handoff model and for the sectorized hexagonal model when some of the messages are delay-sensitive and cannot profit from transmitter or receiver cooperation. For the soft-handoff network we also provide a converse. It matches the multiplexing-gain achieved by our scheme when the multiplexing gain of the delay-sensitive messages is low or moderate or when the cooperation links have high capacities. In these cases, the sum-multiplexing gain is the same as if only delay-tolerant messages (which can profit from cooperation) were sent. A similar conclusion holds for the sectorized hexagonal model, when the capacities of the cooperation links are large. (10.1109/itw.2018.8613499)
  DOI : 10.1109/itw.2018.8613499
• How Memorizing Positions or Directions Affects Gesture Learning?
  
  • Fruchard Bruno
  • Lecolinet Eric
  • Chapuis Olivier
  , 2018, pp.107--114. Various techniques have been proposed to faster command selection. Many of them either rely on directional gestures (e.g. Marking menus) or pointing gestures using a spatially-stable arrangement of items (e.g. FastTap). Both types of techniques are known to leverage memorization, but not necessarily for the same reasons. In this paper, we investigate whether using directions or positions affects gesture learning. Our study shows that, while recall rates are not significantly different, participants used the novice mode more often and spent more time while learning commands with directional gestures, and they also reported more physical and mental efforts. Moreover, this study also highlights the importance of semantic relationships between gestural commands and reports on the memorization strategies that were elaborated by the participants. (10.1145/3279778.3279787)
  DOI : 10.1145/3279778.3279787
• Information theory as a unified tool for understanding and designing human-computer interaction
  
  • Liu Wanyu
  , 2018. Information theory has influenced a large number of scientific fields since its first introduction in 1948. Apart from Fitts' law and Hick's law, which came out when experimental psychologists were still enthusiastic about applying information theory to various areas of psychology, the relation between information theory and human-computer interaction (HCI) has rarely been explored. This thesis strives to bridge the gap between information theory and HCI by taking the stance that human-computer interaction can be considered as a communication process and therefore can be characterized using information-theoretic concepts. The three main contributions are: (1) a detailed historical perspective on how information theory influenced psychology and HCI, particularly an in-depth discussion and analysis of how relevant Hick's law is to HCI; (2) a Bayesian Information Gain (BIG) framework that quantifies the information sent by the user to the computer to express her intention; and (3) a further illustration of the advantages of using information-theoretic measures to evaluate input performance and to characterize the rich aspects of an interaction task. This thesis demonstrates that information theory can be used as a unified tool to understand and design human-computer communication.
• Towards information modeling for a QoS-aware support in the lifecycle of virtual networks
  
  • Diaz Gladys
  • Sibilla Michelle
  • Simoni Noëmie
  , 2019, pp.1-6. Information network modeling is nowadays a popular area of research. Especially, the introduction of virtualization technologies is changing the lifecycle of systems. Virtualization techniques allow distinguishing different levels: applications, networks, and equipment. The design of new virtual networks must consider the requirements of all these levels in a complementary fashion. Quality of Service (QoS) is still the one of the main key feature to be integrated. In this paper, we focus on the virtualized environments where we define a generic concept, called the VirtualElement. Indeed, we are interested in the representation of information enabling the automation of deployment, monitoring and management tasks for virtual networks. For this purpose we characterize the VirtualElement by constrains representing its functional and non-functional behavior. We apply our model at different phases of lifecycle of virtual networks by defining the service profiling. We propose a translation of the virtual network into an OVF file. (10.1109/ATNAC.2018.8615430)
  DOI : 10.1109/ATNAC.2018.8615430
• Towards information modeling for a QoS-aware support in the lifecycle of virtual networks
  
  • Diaz Gladys
  • Sibilla Michelle
  • Simoni Noémie
  , 2018, pp.1-6. Information network modeling is nowadays a popular area of research. Especially, the introduction of virtualization technologies is changing the lifecycle of systems. Virtualization techniques allow distinguishing different levels: applications, networks, and equipment. The design of new virtual networks must consider the requirements of all these levels in a complementary fashion. Quality of Service (QoS) is still the one of the main key feature to be integrated. In this paper, we focus on the virtualized environments where we define a generic concept, called the VirtualElement. Indeed, we are interested in the representation of information enabling the automation of deployment, monitoring and management tasks for virtual networks. For this purpose we characterize the VirtualElement by constrains representing its functional and non-functional behavior. We apply our model at different phases of lifecycle of virtual networks by defining the service profiling. We propose a translation of the virtual network into an OVF file. (10.1109/ATNAC.2018.8615430)
  DOI : 10.1109/ATNAC.2018.8615430
• Set Labelling using Multi-label Classification
  
  • Sanjaya Ngurah Agus
  • Read Jesse
  • Abdessalem Talel
  • Bressan Stéphane
  , 2018, pp.216-220.
• Harnessing Truth Discovery Algorithms On The Topic Labelling Problem
  
  • Sanjaya Ngurah Agus
  • Ba Mouhamadou Lamine
  • Abdessalem Talel
  • Bressan Stéphane
  , 2018, pp.8-14.
• EXAD: A System for Explainable Anomaly Detection on Big Data Traces
  
  • Song Fei
  • Stiegler Arnaud
  • Diao Yanlei
  • Read Jesse
  • Bifet Albert
  , 2018. Big Data systems are producing huge amounts of data in real-time. Finding anomalies in these systems is becoming increasingly important, since it can help to reduce the number of failures, and improve the mean time of recovery. In this work, we present EXAD, a new prototype system for explainable anomaly detection, in particular for detecting and explaining anomalies in time-series data obtained from traces of Apache Spark jobs. Apache Spark has become the most popular software tool for processing Big Data. The new system contains the most well-known approaches to anomaly detection, and a novel generator of artificial traces, that can help the user to understand the different performances of the different methodologies. In this demo, we will show how this new framework works, and how users can benefit of detecting anomalies in an efficient and fast way when dealing with traces of jobs of Big Data systems.
• Towards integrated transmitters for short-reach optical links in the C-band
  
  • Peucheret Christophe
  • Chaibi Mohamed
  • Bramerie Laurent
  • Gay Mathilde
  • Hassan Karim
  • Erasme Didier
  , 2018. In this talk, we will review our recent demonstrations of novel transmitter structures employing Si or III-V integration for inter data center links. We will in particularly look at how the effect of group-velocity dispersion in the C-band can be circumvented in multi-(OFDM) and single-carrier (PAM4) modulation formats thanks to specially designed transmitter architectures.
• Identification des efforts au chevalet d'un instrument cordes. Applications pour la synthèse hybride
  
  • Dujourdy Hugo
  • Ablitzer Frédéric
  • Cabaret Jérémy
  • David Bertrand
  • Gautier François
  , 2018.
• Reliability Emphasized MTJ/CMOS Hybrid Circuit Towards Ultra-Low Power
  
  • Cai Hao
  • Han Menglin
  • Wang You
  • Naviner Lirida
  • Liu Xinning
  • Yang Jun
  • Zhao Weisheng
  , 2018, pp.1-5. (10.1109/DCIS.2018.8681471)
  DOI : 10.1109/DCIS.2018.8681471