Publications

2018

• Toward an analytical model of social-media marketing capabilities
  
  • Kefi Hajer
  • Abdessalem Talel
  • Indra Sitesh
  Management & Data Science, Management & Data Science, 2018, 2 (2). In this research, we adopt a design science approach to develop an analytical framework of social media capabilities of the firm. The approach has been applied in the case of five brands in the cosmetics industry using social network analysis, consumer engagement, descriptive statistics and sentiment analysis. The findings allow us to enlighten the competitive positioning of these brands during the period under study. (10.36863/mds.a.3964)
  DOI : 10.36863/mds.a.3964
• Flexible optical transport networks : benefits of the combination of time and spectral domains to adapt the granularity of optical resources to the needs of operators
  
  • Han Bing
  , 2018. The ever increasing traffic demand is a big challenge for the operators to increase the network capacity while lowering CAPEX and OPEX. To address these challenges, it is expected to increase flexibility of the network resources allocation while realizing directly the IP traffic routing in the optical layer. In this background, the “TIme and Spectral optical Aggregation (TISA)” approach has thus been proposed to allow a purely optical aggregation with the finest possible granularity thanks to the combination of temporal and spectral domains. The main objective of this thesis was to realize a proof of concept of the TISA concept to demonstrate experimentally the feasibility of the TISA solution. To accomplish this objective, we have designed and realized a multi-band OFDM optical burst transmitter with a narrow linewidth fast tunable laser. This laser has been implemented by combining external cavity lasers with semiconductor optical amplifier based optical gates. The transmitter is able to generate OFDM bursts in both DP-QPSK and DP-16QAM modulation formats while introducing a small penalty compared to the continuous flows configuration. We have performed the transmission of these bursts through the TISA network and have measured the bit error rates, which show less than 1 dB penalty for both modulation formats. Our results clearly show the feasibility of the TISA solution and demonstrate the ability to perform a purely optical aggregation/disaggregation and transparent routing while offering a sub-wavelength granularity in both time and spectral domains at the optical layer level.
• Cache-timing Attack Detection and Prevention Application to Crypto Libs and PQC
  
  • Carre Sebastien
  • Facon Adrien
  • Guilley Sylvain
  • Takarabt Sofiane
  • Schaub Alexander
  • Souissi Youssef
  , 2019, pp.13-21. With the publication of Spectre & Meltdown attacks, cache-timing exploitation techniques have received a wealth of attention recently. On the one hand, it is now well understood which some patterns in the C source code create observable unbalances in terms of timing. On the other hand, some practical cache-timing attacks (or Common Vulnerabilities and Exposures) have also been reported. However the exact relationship between vulnerabilities and exploitations is not enough studied as of today. In this article, we put forward a methodology to characterize the leakage induced by a "non-constant-time" construct in the source code. This methodology allows us to recover known attacks and to warn about possible new ones, possibly devastating. (10.1007/978-3-030-16350-1_2)
  DOI : 10.1007/978-3-030-16350-1_2
• Efficient Bayesian Model Selection in PARAFAC via Stochastic Thermodynamic Integration
  
  • Huy Nguyen Thanh
  • Şimşekli Umut
  • Richard Gael
  • Cemgil Ali Taylan
  IEEE Signal Processing Letters, Institute of Electrical and Electronics Engineers, 2018. Parallel factor analysis (PARAFAC) is one of the most popular tensor factorization models. Even though it has proven successful in diverse application fields, the performance of PARAFAC usually hinges up on the rank of the factorization, which is typically specified manually by the practitioner. In this study, we develop a novel parallel and distributed Bayesian model selection technique for rank estimation in large-scale PARAFAC models. The proposed approach integrates ideas from the emerging field of stochastic gradient Markov Chain Monte Carlo, statistical physics, and distributed stochastic optimization. As opposed to the existing methods, which are based on some heuristics, our method has a clear mathematical interpretation, and has significantly lower computational requirements, thanks to data subsampling and parallelization. We provide formal theoretical analysis on the bias induced by the proposed approach. Our experiments on synthetic and large-scale real datasets show that our method is able to find the optimal model order while being significantly faster than the state-of-the-art.
• Lightweight and Wide-Angle Metamaterial Absorbing Material Concept
  
  • Lepage A. C.
  • Pinto Yenny
  • Rance Olivier
  • Begaud Xavier
  • Capet Nicolas
  , 2018.
• Flexibility and dynamicity for open network-as-a-service: From VNF and architecture modeling to deployment
  
  • Boubendir Amina
  • Bertin Emmanuel
  • Simoni Noémie
  , 2018.
• Evolution of LiFePO 4 thin films interphase with electrolyte
  
  • Dupré N.
  • Cuisinier M.
  • Zheng Y.
  • Fernandez V.
  • Hamon J.
  • Hirayama M.
  • Kanno R.
  • Guyomard D.
  Journal of Power Sources, Elsevier, 2018, 382, pp.45 - 55. (10.1016/j.jpowsour.2018.02.029)
  DOI : 10.1016/j.jpowsour.2018.02.029
• Strategic Surveillance Against Primary User Emulation Attacks in Cognitive Radio Networks
  
  • Ta Duc-Tuyen
  • Nguyen-Thanh Nhan
  • Maillé Patrick
  • Nguyen van Tam
  IEEE Transactions on Cognitive Communications and Networking, IEEE, 2018, 4 (3), pp.582-596. Selfish primary user emulation (PUE) is a serious security problem in cognitive radio networks. By emitting emulated incumbent signals, a PUE attacker can selfishly occupy more channels. Consequently, a PUE attacker can prevent other secondary users from accessing radio resources and interfere with nearby primary users. To mitigate the selfish PUE, a surveillance process on occupied channels could be performed. Determining surveillance strategies, particularly in multi-channel context, is necessary for ensuring network operation fairness. Since a rational attacker can learn to adapt to the surveillance strategy, the question is how to formulate an appropriate modeling of the strategic interaction between a defender and an attacker. In this paper, we study the commitment model in which the network manager takes the leadership role by committing to its surveillance strategy and forces the attacker to follow the committed strategy. The relevant strategy is analyzed through the Strong Stackelberg Equilibrium (SSE). Analytical and numerical results suggest that, by playing the SSE strategy, the network manager significantly improves its utility with respect to playing a Nash equilibrium (NE) strategy, hence obtains a better protection against selfish PUEs. Moreover, the computational effort to compute the SSE strategy is lower than to find a NE strategy. (10.1109/TCCN.2018.2826552)
  DOI : 10.1109/TCCN.2018.2826552
• An Online Approach to D2D Trajectory Utility Maximization Problem
  
  • Singh Bedi Amrit
  • Rejawat Ketan
  • Coupechoux Marceau
  , 2018, pp.1-9. The emergence of social media and the associated mobile applications has ushered a culture of constant connectivity. The cybercitizens of today are increasingly willing to modify their behavior in order to stay connected. The current paper formulates the trajectory optimization problem for commuting users communicating through a device-to-device (D2D) link. We consider a pair of pedestrians seeking to reach their respective destinations, while using the D2D link for data exchange applications such as file transfer, video calling, and online gaming. In order to enable better D2D connectivity, the pedestrians are willing to deviate from their respective shortest paths, at the cost of reaching their destinations slightly late. A generic trajectory optimization problem is formulated and solved for the case when full information about the problem in known in advance. Motivated by the D2D user's need to keep their destinations private, we also formulate a regularized variant of the problem that can be used to develop a fully online algorithm, that is shown to achieve a sublinear offline regret and satisfy the user mobility constraints. The theoretical results are backed by detailed numerical tests that establish the efficacy of the proposed algorithms under various settings.
• The Perils of Confounding Factors: How Fitts' Law Experiments can Lead to False Conclusions
  
  • Gori Julien
  • Rioul Olivier
  • Guiard Yves
  • Beaudouin-Lafon Michel
  , 2018 (196), pp.1-10. The design of Fitts' historical reciprocal tapping experiment gravely confounds index of difficulty ID with target distance D: Summary statistics for the candidate Fitts model and a competing model may appear identical, and the validity of Fitts' model for some tasks can be legitimately questioned. We show that the contamination of ID by either target distance D or width W is due to the common practices of pooling and averaging data belonging to different distance-width (D,W) pairs for the same ID, and taking a geometric progression for values of D and W. We analyze a case study of the validation of Fitts' law in eye-gaze movements, where an unfortunate experimental design has misled researchers into believing that eye-gaze movements are not ballistic. We then provide simple guidelines to prevent confounds: Practitioners should carefully design the experimental conditions of (D,W), fully distinguish data acquired for different conditions, and put less emphasis on r 2 scores. We also recommend investigating the use of stochastic sampling for D and W. (10.1145/3173574.3173770)
  DOI : 10.1145/3173574.3173770
• Future trends in optical networks—and why you should care
  
  • Ware Cédric
  , 2018.
• Introduction to special issue: ACM SIGMETRICS 2016
  
  • Bonald Thomas
  • Ganesh Ayalvadi
  Queueing Systems, Springer Verlag, 2018, pp.1-2.
• Batched packet processing for high-speed software data plane functions
  
  • Barach David
  • Linguaglossa Leonardo
  • Damjan Marion
  • Pfister Pierre
  • Pontarelli Salvatore
  • Rossi Dario
  • Tollet Jerome
  , 2018.
• Failure of the current modulation driven linewidth broadening factor for analyzing the optical linewidth behavior of quantum dot lasers
  
  • Huang Heming
  • Duan Jianan
  • Lu Zhaoyang
  • Poole Philip
  • Grillot Frédéric
  , 2018.
• Recent advances in InAs/GaAs quantum dot lasers with short optical feedback
  
  • Grillot Frédéric
  • Huang Heming
  • Lin Lyu-Chih
  • Chen Chih-Ying
  • Arsenijevi Dejan
  • Lin Fan-Yi
  • Bimberg Dieter
  , 2018.
• Self-awareness and Decision-taking in Socio-Cyber-Physical Systems. An Architectural Perspective
  
  • Diaconescu Ada
  • Pitt Jeremy
  , 2018.
• Greedy stochastic algorithms for entropy-regularized optimal transport problems
  
  • Abid Brahim Khalil
  • Gower Robert M.
  , 2018.
• Dynamic Local Models for Online Recommendation
  
  • Al-Ghossein Marie
  • Abdessalem Talel
  • Barré Anthony
  , 2018, pp.1419-1423. (10.1145/3184558.3191586)
  DOI : 10.1145/3184558.3191586
• Digital vs. analog coherent combining on RL-ESPAR antennas
  
  • Bucheli Garcia Juan
  • Sibille Alain
  • Kamoun Mohamed
  , 2018. The reactively loaded parasitic array radiator (RLESPAR) receiver has been acknowledged due to its compactness, fabrication cost and reconfigurability; showing significant beamforming trade-off capabilities compared to conventional multielement receiver ends. When used on reception, one relevant problem is to find the value of the reactance loadings to obtain the best signal to noise ratio performance. In the current paper, we compare the performance of digital and analog coherent combining over RL-ESPAR. Particularly, digital combining is realized by virtually rotating such a receiver and then applying the known maximum ratio combining (MRC) technique. On the other hand, analog combining is realized by synthesizing MRC on the reactance loadings via a technique that relies on the simultaneous perturbation stochastic approximation method(SPSA) method (widely applied on ESPAR in the literature). We show that analog combining exceeds digital combining and single dipole reception by around 3 dBs and 4 dBs via the proposed technique, respectively.
• Multistage Single Clad 2μm TDFA with a Shared L-Band Pump Source
  
  • Tench Robert
  • Romano Clément
  • Delavaux Jean-Marc
  , 2018, pp.paper 10654-31.
• An UWB Coplanar Waveguide Fed Integrated IFA Design for Wearable Communications
  
  • Du Jinxin
  • Roblin Christophe
  , 2018. An UWB CPW-fed Integrated “IFA-like” design is proposed. The new design has a much longer short-ended arm and a much narrower gap between the radiating arms and the coplanar ground plane. The bandwidth increase is achieved thanks to the superposition of multi-resonances induced by the joint effects of a strong coupling between the radiating arms and the ground plane and the excitation of both even and odd CPW modes in the feedline. The design has been tested for two types of substrate, FR4 and denim, for which –10 dB bandwidths of more than 60 % and 87 % are achieved respectively. Body proximity effects are analysed for the denim IIFA by simulation. It shows that the antenna remains working well in the presence of human body thanks to its UWB feature. This extremely simple, lowprofile and easily integrable UWB IIFA-like design can be used for wearable communications (on-body/off-body), supporting multi-standards.
• Are All People Married? Determining Obligatory Attributes in Knowledge Bases
  
  • Lajus Jonathan
  • Suchanek Fabian M.
  , 2018.
• A secure IoT architecture for streaming data analysis and anomaly detection
  
  • Clémençon Stéphan
  • Boudabous Safa
  • Jelassi Ons
  • Roca Mariona Caros
  , 2018. Discovery of repeating temporal patterns and prediction based on time stamped data generated by IoT services raise important methodological issues. A typical predictive problem, addressed in this paper, consists in the early detection of change points or anomalies, that may be caused by a malicious use of the system for instance. Although online anomaly detection is now the subject of much attention in the data science literature, motivated by crucial industrial applications such as predictive maintenance and health monitoring of complex infrastructures, the rapidly changing environment inherent to most IoT applications makes this task even more challenging. Beyond the crucial control of the false alarm rate, the quasi real-time analysis must take into account the efficiency of computing resources and the possible security risks in data transfers over the network. We propose here an architecture for analyzing IoT datastreams data and a dedicated method for on-line anomaly/novelty detection, based on nonparametric (mean discrepancy) test statistics and multiple hypothesis testing techniques. Numerical results based on experiments involving synthetic datastreams and real energy consumption datastreams provides empirical evidence of the relevance of the methodology proposed.
• Output Fisher Embedding Regression
  
  • Djerrab Moussab
  • Garcia Rojas A.
  • Sangnier Maxime
  • d'Alché-Buc Florence
  Machine learning Journal, 2018, 108. We investigate the use of Fisher vector representations in the output space in the context of structured and multiple output prediction. A novel, general and versatile method called "Output Fisher Embedding Regression" (OFER) is introduced. Based on a probabilistic modeling of training output data and the minimization of a Fisher loss, it requires to solve a pre-image problem in the prediction phase. For Gaussian Mixture Models and State-Space Models, we show that the pre-image problem enjoys a closed-form solution with an appropriate choice of the embedding. Numerical experiments on a wide variety of tasks (time series prediction, multi-output regression and multi-class classification) highlight the relevance of the approach for learning under limited supervision like learning with a handful of data per label and weakly supervised learning.
• Quantum Technologies : Impact on Cybersecurity (and AI)
  
  • Alleaume Romain
  , 2018.