Archives des séminaires

Amphi OPALE, 14H, Télécom ParisTech, 46 rue barrault, Paris 13

"Attaques par injection de fautes et contre-mesures : passé, présent, futur"

Les attaques par injection de fautes sont des techniques extrêmement puissantes pour extraire des secrets d'un circuit intégré. Les toutes premières contre-mesures, développées il y a une vingtaine d’années, ont posé les bases des stratégies de protection. Ce séminaire débutera par leur présentation, et nous vous en proposerons une classification. Une comparaison du coût de chaque contre-mesure sera également réalisée, ainsi qu'une analyse de leur niveau de sécurité, vis-à-vis des menaces existantes lors de leur publication, mais également vis-à-vis d'autres menaces plus récentes.

Ces dernières années, les perturbations électromagnétiques intentionnelles ont suscité un grand intérêt comme moyen d'injection de fautes, à la fois pour des aspects pratiques mais surtout pour leur potentiel à contourner certaines stratégies de protection. Il en découle un besoin de comprendre finement l'impact de telles injections au sein des circuits intégrés. Cependant, les méthodes de caractérisation et de modélisation de l'état de l'art se sont avérées incomplètes, et nous vous exposerons dans la seconde partie de ce séminaire les améliorations que nous y avons apportées, et les résultats ainsi obtenus.

in A301 at 1.30pm

Abstract :

Physically unclonable functions (PUF) are used in various applications
requiring robust authentication. These systems exploit unpredictable
process variations in electronic circuits. These process variations
uniquely identify the produced hardware, which exhibit distinct
properties in terms, for example, of delay propagations inside the
circuit. By measuring and exploiting these properties, one can determine
a "fingerprint" of the circuit, which can not be physically replicated.
This fingerprint can then be used, for instance, to produce a
cryptographic key. The advantage is that this key does not need to be
explicitly stored, which reduces the security risk. Other applications
include challenge-response protocols, where the responses are determined
from the physical properties of the circuit.

For a given type of PUF, the Loop-PUF, these delay propagation
differences can be modeled by n Gaussian random variables. A challenge
corresponds to a vector of +/- 1 values, and an identifier bit is the
sign of the signed sum of the Gaussian realizations, with signs
corresponding to those of the challenge vector. We try to adress the
following question: what is the joint entropy of these sign bits ?

The exact calculation of the maximum entropy, when considering the set
of all possible challenges, can be carried out only for very small
values of n. We provide a combinatorial extension that provides the
exact values for n = 3 and 4. For n greater or equal to 5, the method
soon becomes intractable and one has recourse to numerical computations.
The value of the maximum entropy can be estimated reliably by defining
equivalence classes of challenges corresponding to the same value of
joint probabilities. This method was found to be numerically tractable
for values of n up to 7. Asymptotic expressions for the max-entropy are
found using the theory of threshold boolean functions.

Amphi JADE, 14H, Télécom ParisTech, 46 rue barrault, Paris 13

""Métrologie des fréquences optiques et peignes de fréquences compacts""

Les peignes de fréquences compacts sont des outils métrologiques ayant vocation à être intégrés dans des systèmes embarqués dédiés entre autres à la génération d’horloges pour des applications spatiales, à la spectroscopie de précision et aux tests fondamentaux nécessitant des références de fréquences, à la génération d’ondes optiques/micrométriques/Terahertz ultra-stables pour des applications en télécommunications (radar, communication sans fil, communications optiques cohérentes, etc.).

Lors du séminaire je présenterai dans un premier temps les grands principes de la métrologie Temps-Fréquences : les outils expérimentaux (référence, cavité de transfert) et les outils mathématiques (variance d’Allan) nécessaire à la caractérisation métrologique des oscillateurs optiques, et ce, au travers de deux exemples de techniques de stabilisation en fréquence d’un laser : l’une sur une cavité ultra-stable (Pound-Drever-Hall) et l’autre sur une transition moléculaire par la technique d’absorption saturée en cellule.

Dans un second temps, je présenterai mon activité de recherche sur les peignes de fréquences compacts et les résultats de stabilisation et de référencement de lasers à semi-conducteurs à blocage de modes optiques passifs, sur une cavité de transfert à fibre référencée sur une transition d’Acétylène détectée en absorption saturée. Les avantages d’une stabilisation mixte conjuguant stabilisation optoélectronique et affinement du peigne par injection optique seront discutés.

at 11am in F900,

Abstract:
We consider transmission over a cloud radio access network (CRAN) focusing
on the framework of oblivious processing at the relay nodes (radio units),
i.e., the relays are not cognizant of the users' codebooks.
This approach is motivated by future wireless communications
(5G and beyond) and the theoretical results connect to a variety
of different information theoretic models and problems.
First it is shown that relaying a-la Cover-El Gamal, i.e.,
compress-and-forward with joint decompression and
decoding, which reflects 'noisy network coding,' is optimal.
The penalty of obliviousness is also demonstrated to be
at most a constant gap, when compared to cut-set bounds.
Naturally, due to the oblivious (nomadic) constraint the CRAN problem
intimately connects to Chief Executive Officer (CEO) source(s) coding
under a logarithmic loss distortion measure.
Furthermore, we identify and elaborate on some interesting
connections with the distributed information bottleneck model for which we
characterize optimal tradeoffs between rates (i.e., complexity) and
information (i.e., accuracy) in the discrete and vector Gaussian frameworks.
Further connections to 'information combining' and 'common reconstruction'
are also pointed out. In the concluding outlook, some interesting problems
are mentioned such as the characterization of the optimal input distributions
under users' power limitations and rate-constrained compression at the
relay nodes,

---------------------------------------------------------------------------
Joint work with: I.E. Aguerri (Paris Research Center, Huawei France)
A. Zaidi (Universite Paris-Est, Paris) and G. Caire (USC-LA and TUB, Berlin)
The research is supported by the European Union's Horizon 2020 Research And
Innovation Programme: no. 694630.

tomb (Linux only) is a cryptographic utility based on the Linux Unified Key Setup (LUKS) 

standard and the disk encryption subsystem of Linux kernel (dm-crypt device mapper). 

It can be used to encrypt directories, turning them into binary files. 

It is very well designed and easy to use. pass is a password manager with many interesting 

characteristics. It is open source, does not rely on third party (more or less trustable) 

servers, integrated with git and tomb, supported by Firefox / Chrome... extensions, etc. 

This seminar will briefly present these two tools and how they can be jointly used for 

the highest security of your personal data. Then, we will demonstrate them by installing 

all needed components, generating the keys and use them for real use cases.

Amphi B310, 14H, Télécom ParisTech, 46 rue barrault, Paris 13

"Spatial Data Focusing: an alternative to Beamforming for geocasting scenarios"

The capability of an antenna to focus radiated signals into a well defined direction is fundamentally limited by its size (the smaller, the less directive), as the result of diffraction or, equivalently, owing to the properties of the Fourier transform. This applies to single antennas as well as to arrays of multiple antennas.

In this seminar, a Spatial Data Focusing technique is introduced as an alternative scenario to overcome antenna array's beamwidth limitations due to the finite aperture size. The proposed approach aims to focus the transmitted data rather than the transmitted power. This scheme enables wireless broadcast of information to specific spatial locations, using fewer antenna elements compared to classical beamforming techniques. Different configurations will be discussed to implement this scheme and it will be shown that focusing the data is spatially more selective than focusing the power.

in A301 at 10am

Abstract

5G Systems are expected to have a 1000-fold increase in mobile data traffic compared to 2015, largely increased mobile devices per unit area, and decreased latency, especially for M2M/D2D communications. Among the front-runner technologies to achieve such lofty aims are multi-tier heterogenous networks that combine outdoor RF macrocells and indoor gigabit (wide-band) small-cells, and massive MIMO. These technologies include very high carrier frequencies with massive bandwidths, extreme base station and device densities and unprecedented numbers of antennas.
In this talk, I will go over the work done in my group at the University of Waterloo in signal processing and algorithms design and analysis for 5G networks. The talk is divided into three main parts. First, I will briefly talk about hybrid beamforming for single and multi-user massive MIMO.
Second, I will present a brief analysis of some HetNets from the physical layer perspectives (i.e., outage probability and diversity/degrees of freedom analysis). Lastly, I will discuss techniques to simplify the detection (combined with decoding) algorithms at the receiver side.
Massive MIMO are used primarily to combat the high absorption rate of the mm-wave channels; however, this comes with a price: namely a large number of (expensive) RF chains (i.e., DAC and PA) as well as a huge overhead for training and channel knowledge feedback. Hybrid analog (RF) and digital (baseband) beamforming allows one to reduce the required number of RF chains as well as to have a limited feedback. The appropriate choice of beamforming codebooks combined with the selection algorithms can allow such simplification with a limited loss compared to the optimal solution.
Conventionally, HetNets are adopted to broaden the network coverage and alleviate the near-far problem in cellular networks.
In the literature, the cooperative benefits of HetNets are limited to load balancing and traffic offloading, or sharing resources. I will talk about the cooperation benefits in HetNets in terms of outage probability and diversity order. Finally, I will talk about how to reduce the complexity of near-optimal detection algorithms at the receiver side using established techniques such as lattice reduction and conditional optimization.

Bio: Mohamed Oussama Damen is an Electrical and Computer Engineering Professor at the University of Waterloo.
Professor Damen has an extensive background in research positions at multiple academic institutions including École Nationale Supérieure des Télécommunications in Paris, France; the University of Minnesota and the University of Alberta. In June of 2004, he joined the University of Waterloo, where he then became the Nortel Networks Associate Chair in Advanced Telecommunications from April 2005 to April 2010.
His current research interests include coding theory (particularly regarding lattices, coding and decoding algorithms), cross-layer optimization, multiple-input multiple-output and space time communications, multiuser detection, and wireless communications.
Professor Damen has received several awards including the University of Waterloo ECE Research Excellence Award in 2007, Early Researcher Award in the Province of Ontario for 2007 to 2010 and the Junior Research Fellowship from the French Research Ministry in 1996 to 1999. Professor Damen has published numerous articles and journals with, and is currently a senior level member of, IEEE.

Amphi B312, 14H, Télécom ParisTech, 46 rue barrault, Paris 13

Le besoin en systèmes de mesure fonctionnant en totale autonomie existe depuis longtemps, mais ne pouvait être couvert jusqu’à présent que par des systèmes interrogeables sur place. Les données acquises n’étaient de ce fait pas toujours actualisées et ne donnaient qu’un instantané de l’environnement exploré plutôt qu’une surveillance permanente de celui-ci.
L'arrivée de véhicules autonomes intégrant des capteurs de technologies modernes (LIDAR, Radar, vidéo, etc.) ouvre de nouvelles perspectives dans ce domaine. Ces équipements constituent des moyens de plus en plus sophistiqués pour acquérir des informations de toute nature afin d’explorer un environnement donné. L'utilisation de capteurs couplée à un système autonome permet de réaliser une mission sans intervention externe. Cette approche permet de s'affranchir des contraintes d'infrastructure existantes ou de communication difficile comme par exemple des environnements présentant des infrastructures en panne ou détruites telles que les situations faisant suite à des accidents ou des catastrophes. Les informations collectées ont une finalité duale : d’une part, comprendre et modéliser l'environnement pour une bonne réalisation de la mission et d’autre part, réutiliser cette information dans un cadre plus large d'aide à la décision
Les données traitées sont produites par un ensemble de capteurs variés et déployés en réseaux pour la collecte en temps réel. Les données produites sont de divers types : données de distance obtenues par ders capteurs ultrasonores (temps de vol), données de distance obtenues par des capteurs optiques laser (LIDAR), données de position et d'attitude produites par les systèmes inertiels (accéléromètres, magnétomètres, gyromètres, etc.), données d'odométrie, données d'environnement telles que la température, la pression, etc.
Les difficultés résident d’abord dans l'utilisation de processeurs de basse puissance (capacité de calcul faible) pour des raisons d’optimisation énergétique  dans un contexte de système embarqué critique et de prise en compte de l’incertitude des données. Cette difficulté intervient aussi bien en ce qui concerne la réduction des données acquises, leur traitement, que leur protection par des mécanismes logiques et cryptographiques.
Une autre question intéressante porte sur l’utilisation des données acquises : dans le cas où ces données doivent être protégées par des techniques cryptographiques, comment les rendre utilisables par le système autonome pour sa propre navigation tout en les protégeant contre des fuites intempestives ?
Après un état des difficultés et contraintes, plusieurs pistes de résolution seront abordées. Enfin, l'état actuel du projet et des divers prototypes actuel seront présentés.

at 2.30pm in A301

Abstract:

The main goal of future cellular communication systems is to handle high data rate requirement of users with a reasonable latency. There are two main limiting factors against the high data rate communication: Fading and interference. While fading reduces the coverage area and reliability of any point-to-point connection, e.g. mobile user to base station (BS), interference puts constraint on reusability of spectral resources (frequency slots, time etc.) in space, hence restricting the overall spectral efficiency expressed in bits/sec/Hz/BS. Joint processing of the signals received by multiple BS’s, which enables the exploitation of correlation between received signals, is one of the methods to reduce the detrimental effect of interference. In this talk, I will present coding schemes that enable the joint processing such that one of them is based on cooperation between BS’s and the other two are based on employment of cloud random access networks (C-RAN). I will also present lower bounds on the achievable degrees of freedom (DoF).

Amphi Opale, 14H, Télécom ParisTech, 46 rue barrault, Paris 13

The operating principle of RFID has been observed by many scientists during the last century. Among these observations, two are very relevant to modern RFID: a spying “microphone” called “The Thing” developed by L. Theremin and offered by the Soviet Union to the U.S. Ambassador on August 4, 1945 ; an IRE publication entitled “communication by means of reflected power”  published by H. Stockman for the first time in 1948. These two pioneering contributions demonstrate the effectiveness of RFID technology with its two branches: conventional RFID based on ASICs and chipless RFID, a kind of RF barcode.

Nowadays, with the rapid implementation of the concept of the internet of things which involves tens of billions of items to be connected, the development of low cost, low power, wireless and recyclable devices for identification and sensing is highly needed. chipless RFID seems to be an attractive solution for that purpose.

This communication is dedicated to cover chipless RFID technology, from operating principle and technology development to applications scenarios and market expectations. After a brief history of RFID, the talk will focus on chipless technology, the principle of information coding in chipless tags with several examples. Then the transformation of chipless tags into sensors and their capabilities in the context of IOT will be discussed. The presentation will also cover the impact of RF regulations on the performance of chipless solutions. Finally, some   advanced examples and remarks will conclude the presentation.

Docker, Git, GitLab and the GitLab Continuous Integration, application to automatic evaluation of students works. 

at 1pm in A301

Abstract:

As the wireless mobile communications are witnessing unprecedented growth, the future 5G networks will have to satisfy the demands on high-volume data traffic and at the same time strong requirements on the latency, error rates and energy consumption. Therefore, improving the computing capabilities of mobile devices and extend their battery lives are rising among the key challenges for next generation wireless network. In this work, energy harvesting and mobile cloud computing are exploited as two promising technologies to tackle these challenges by designing optimal policies for resource scheduling and computation offloading.

at 11am in Amphi Opale

Abstract

This presentation aims at giving an overview of 5G from a PHY/MAC perspective (with emphasis on the differences with LTE), notably, in terms of enablers and design principles (forward compatibility). The presentation follows what has been standardized for 3GPP New Radio Rel. 15.

The outline of the presentation is as follows:
 
Introduction
·         Background (standardization process, requirements/levers, LTE vs 5G)
Part I: 5G PHY/MAC Enablers
·         Physical channels, physical reference signals
·         Frame structure/numerology
·         Waveform
·         Massive MIMO
·         Synchronization
·         Beam management
Part II: 5G Design principles
·         Forward compatibility
·         Lean design
·         Stay in the box
·         Avoid strict timing relations
·         TDD and FDD design
·         Low latency
Conclusion

at 11am in A301

Biography: 
Navin Kashyap received the B.Tech. degree in Electrical Engineering from the Indian Institute of Technology, Bombay, in 1995, the M.S. degree in Electrical Engineering from the University of Missouri-Rolla in 1997, and the M.S. degree in Mathematics and the Ph.D. degree in Electrical Engineering from the University of Michigan, Ann Arbor, in 2001. From November 2001 to November 2003, he was a postdoctoral research associate at the University of California, San Diego. From 2004 to 2010, he was on the faculty of the Department of Mathematics and Statistics at Queen's University, Kingston, Ontario, Canada. In January 2011, he joined the Department of Electrical Communication Engineering at the Indian Institute of Science, where is currently a Professor. His research interests lie primarily in the application of combinatorial and probabilistic methods in information and coding theory. Prof. Kashyap served on the editorial board of the IEEE Transactions on Information Theory during the period 2009-2014. He is at present an Associate Editor for the SIAM Journal on Discrete Mathematics. He has been appointed as a Distinguished Lecturer of the IEEE Information Theory Society for 2017-2018.

at 2pm in Amphi Opale

Abstract:

Users of cloud systems demand that their data be reliably stored and quickly accessible. Cloud providers today strive to meet these demands through over-provisioning: keeping processors ready to go at all times  and replicating data over multiple servers. Special erasure codes have been designed and adopted in practice as a more storage-efficient way to provide reliability. We will show how coding reduces download time of large files, in addition to providing reliability against disk failures. We will introduce a fork-join queuing framework to model multiple users requesting their data simultaneously, and demonstrate the trade-off between the download time and the amount of storage space. We will explain how for the same total storage used, coding exploits the diversity and parallelism in the system better than today's replication schemes, and hence gives faster download.  At the end, we will mention several problems that arise in distributed computing systems when some servers are straggling in completing their tasks, and the cloud data is hot, large, changing, and expanding.

Bio:

Emina Soljanin is a Professor at Rutgers University. Before moving to Rutgers in January 2016, she was a (Distinguished) Member of Technical Staff for 21 years in the Mathematical Sciences Research of Bell Labs. She works as an information, coding, and, more recently, queueing theorist. Her interests and expertise are wide. Over the past quarter of the century, she has participated in numerous research and business projects, as diverse as power system optimization, magnetic recording, color space quantization, hybrid ARQ, network coding, data and network security, and quantum information theory and networking. Dr. Soljanin served as the Associate Editor for Coding Techniques, for the IEEE Transactions on Information Theory, on the Information Theory Society Board of Governors, and in various roles on other journal editorial boards and conference program committees. She is a 2017 outstanding alumnus of the Texas A&M School of Engineering, an IEEE Fellow, a 2016/17 Distinguished Lecturer for the IEEE Information Theory Society,  and is currently serving as the Second Vice President for the society.

Amphi B312, Télécom ParisTech, 46 rue barrault, Paris 13

Il y a moins de 10 ans, Bell Labs mettait au point les réseaux optiques cohérents, le nouveau standard de l’industrie aux milliards d’utilisateurs. Ils nous permettent d’intensifier l’usage que nous faisons de nos smartphones à coût constant ou parfois même à coût réduit, en continuant de vivre avec l’illusion de communications sans fil (et surtout sans fibre).

Derrière cette révolution, une autre se prépare. Il y a fort à parier que les bénéfices de la numérisation introduite dans les systèmes cohérents ne se limiteront pas à soutenir la croissance des flux de données (numériques, bien sûr). Trop d’objets de notre quotidien ont changé de nature grâce aux technologies numériques. Et si les réseaux numériques numérisés devenaient capables d’écouter, de s’analyser, de prendre des décisions seuls… de devenir intelligents? Certains envisagent aussi de profiter de ces avancées technologiques pour inventer des systèmes de communications par laser entre la terre et l’espace à très haut débit.

Dans les systèmes et les réseaux à fibre optique, les technologies cohérentes ont déjà permis de rendre techniquement et économiquement viables quelques rêves que d’aucuns avaient condamnés à l’oubli. Il y a fort à parier que ces rêves devenus réalités ne seront pas les derniers.

Amphi B310, 16H30, Télécom ParisTech, 46 rue barrault, Paris 13

Supporting IoT communications in future 5G networks is a fundamental challenge. IoT services will rely on the deployment of billions of things communicating various traffic under many different quality of service requirements. Reliability and delay constraints will deeply modify the classical paradigm of communication theory, especially for services generating bursty traffic. For some kind of services often referred to as tactile internet or mission critical communications, where IoT nodes are deployed in smart environments (e.g. smart cities) the communication link is a part of a distributed controlled system and therefore require strong reliability and delay constraints.

After an overview of the main challenges and upcoming technologies, we will focus on scenarios where a very large number of nodes are willing to transmit rare but  reliable short information quantities. We address the problem from an information theory perspective, to derive some fundamental limits associated to this scenario.

at 1pm in A301

Abstract

The conventional single mode Fibers (SMFs) cannot keep up with the exponential growth in the demand for network bandwidth since it almost reached the limitation induced by non-linear Kerr effects.  Spatial Division multiplexing (SDM) is the last degree of freedom can be applied in the optical fiber. SDM is approached through multi-mode fibers (MMFs) or multi-core fibers (MCFs). My thesis focuses on evaluating several MCF configurations and mitigating the impairments that face the MCFs such as Core Dependent Loss (CDL).  We propose Multi-Core Fiber channel model for different systems (without scrambling and with scrambling). This solution allows estimating the CDL so predict the system performance. In addition, a new deterministic scrambling technique is proposed in order to reduce the CDL optimally by installing less number of scramblers.

Amphi B310, Télécom ParisTech, 46 rue barrault, Paris 13

Les réseaux de capteurs corporels représentent un enjeu sociétal important en permettant des soins de meilleure qualité avec un coût réduit. Ces derniers sont utilisés pour détecter des anomalies dès leur apparition et ainsi intervenir au plus vite. Les capteurs sont soumis à de nombreuses contraintes de fiabilité, robustesse, taille et consommation.

Dans cette présentation, les différentes opérations réalisées par les réseaux de capteurs corporels sont analysées. La consommation de chacune d'elles est évaluée afin de guider les axes de recherche pour améliorer l'autonomie énergétique des capteurs. Un capteur pour la détection d'arythmie sur des signaux cardiaques est proposé. Il intègre un traitement du signal via l'utilisation d'un réseau de neurone à cliques. Le système proposé est simulé et offre une exactitude de classification de 95 % pour la détection de trois types d'arythmie. Le prototypage du système via la fabrication d'un circuit mixte analogique/numérique en CMOS 65 nm montre une consommation du capteur de l'ordre de 1,4 mJ.

Pour réduire encore plus l'énergie, une nouvelle méthode d'acquisition est utilisée. Une architecture de convertisseur est proposée pour l'acquisition et le traitement de signaux cardiaques. Cette dernière laisse espérer une consommation de l'ordre de 1,18 nJ pour acquérir les paramètres tout en offrant une exactitude de classification proche de 98 %. Cette étude permet d'ouvrir la voie vers la mise en place de capteurs très basse consommation pouvant durer toute une vie avec une simple pile.

Amphi B310, Télécom ParisTech, 46 rue barrault, Paris 13

Les réseaux de capteurs corporels représentent un enjeu sociétal important en permettant des soins de meilleure qualité avec un coût réduit. Ces derniers sont utilisés pour détecter des anomalies dès leur apparition et ainsi intervenir au plus vite. Les capteurs sont soumis à de nombreuses contraintes de fiabilité, robustesse, taille et consommation.

Dans cette présentation, les différentes opérations réalisées par les réseaux de capteurs corporels sont analysées. La consommation de chacune d'elles est évaluée afin de guider les axes de recherche pour améliorer l'autonomie énergétique des capteurs. Un capteur pour la détection d'arythmie sur des signaux cardiaques est proposé. Il intègre un traitement du signal via l'utilisation d'un réseau de neurone à cliques. Le système proposé est simulé et offre une exactitude de classification de 95 % pour la détection de trois types d'arythmie. Le prototypage du système via la fabrication d'un circuit mixte analogique/numérique en CMOS 65 nm montre une consommation du capteur de l'ordre de 1,4 mJ.

Pour réduire encore plus l'énergie, une nouvelle méthode d'acquisition est utilisée. Une architecture de convertisseur est proposée pour l'acquisition et le traitement de signaux cardiaques. Cette dernière laisse espérer une consommation de l'ordre de 1,18 nJ pour acquérir les paramètres tout en offrant une exactitude de classification proche de 98 %. Cette étude permet d'ouvrir la voie vers la mise en place de capteurs très basse consommation pouvant durer toute une vie avec une simple pile.

Power MOSFET devices are a primary component of power converters, and so are widely used in electronic systems. This widespread usage makes the study of the devices an important domain. Understanding the way these devices function enables to design more performant systems. Also, due to their extensive use, ensuring the security of embedded system involves ensuring the authenticity of the power MOSFET devices they contain. The authenticity must be ensured because a refurbished or a counterfeit device could affect performances and/or security of critical systems such as automobiles.

To achieve device authentication we:
- Designed an automatic measurement system to measure more easily the devices, and
- Used Machine Learning on the measurement curves obtained in order to identify devices

Potential of Optical Wireless Technology in Addressing Spectrum Requirements in Future Smart environments

Jeudi 1er février 2018, 14 H, Amphi Emeraude, télécom ParisTech, 46 rue Barrault, Paris 13

Optical wireless communications (OWC) in the infrared and visible spectra have recently received particular attention due to their potential in providing very high-rate data transmission in various indoor and outdoor application scenarios. This seminar provides an overview on this transmission technology and the challenges in the deployment of such links. We start by briefly presenting the principles of free-space optical communications (FSO) in outdoor applications and reviewing their most important practical limitations. We then consider the emerging field of visible-light communications (VLC) and explain their potential in enabling wireless connectivity in indoor and underwater applications. For the former context (e.g. in smart home applications), the main challenge concerns the limited modulation bandwidth and dynamic range of the commercial LEDs, which are supposed to replace the traditional luminaries in the near future. In the underwater context, on the other hand, the main limitation arises from the high attenuation of the aquatic channel, which seriously affects the attainable transmission range. We discuss the other factors affecting the data transmission rate and the link robustness, and the existing implementation challenges.

in C46 at 11am

Abstract:

There are 99,270,589,265,934,370,305,785,861,242,880 8-variable bent functions. Current construction methods only explain about 1 in a million of these. I will give a gentle introduction to bent functions and difference sets, and I will describe a construction method from difference sets that can be applied to bent functions to achieve “new” 8-variable bent functions (bent functions that don’t currently have a known construction method). We will conclude with many open questions

Bio:

Jim Davis, Professor of Mathematics at the University of Richmond since 1988, does research in combinatorics and error correcting codes. He spent two years working for Hewlett-Packard and he has 15 patents stemming from that work. He has published more than 50 papers, including several with undergraduates as coauthors (one of those also has Sihem Mesnager as coauthor). Outside of mathematics, he is an avid squash and bridge player.

in A301 at 1.30pm

Abstract :

My research lies in Human-Computer Interaction (HCI). In my thesis, I design interaction techniques and study user performance from an information-theoretic perspective, by capturing the uncertainty in the communication process and exploring the notion of combining human intelligence with machine power. Using the concepts from Information Theory and Bayesian Experimental Design, I propose BIG (Bayesian Information Gain), a framework to quantify the information sent by the user to the computer to express her intention. Two applications, BIGnav for multiscale navigation and BIGFile for hierarchical file retrieval, demonstrate how the computer can play a more active role and work together with the user to achieve shared goals. I also explore the concept of user information capacity to inform generalizable interaction designs.

Maritime space is one of the important means of international transportation. This environment is known to be a dynamic spatial space where disruptions exist. In this report we propose a model to measure and evaluate the effect of these disruptions on the network.

To achieve this, multi-agent system is used, where different components of the maritime network are represented as agents in a shared environment. Using an agent-based system, we will simulate vessels trajectories and spatial-temporal features of disruption to assess the vulnerability of the maritime network.

14H, Amphi B312, Télécom paristech, 46 rue Barrault, Paris 13

Semi-conductor lasers invented in 1962 are vital to our modern daily life. For example, they generate the optical impulses that carry ever-greater amounts of information in fiber-optic networks over great distances. The emergence of irregular and unpredictable pulsations and dynamical instabilities from a laser were first noted during the very early stages of the development of lasers. Pulses with amplitude varying in an erratic manner were reported in the output of the ruby solid-state laser. However, the lack of knowledge of what would later be termed butterfly effect i.e. deterministic chaos resulted in these initial observations being either left unexplained or wrongly attributed to noise.

This presentation will highlight the fundamental physics underpinning the butterfly effect in semiconductor lasers and also the opportunities in harnessing it for potential applications. The availability and ease of operation of semiconductor lasers, in a wide range of configurations, make them a convenient test-bed for exploring basic aspects of nonlinear and chaotic dynamics. Avenues for future research and development of quantum cascade lasers and nanoscale devices will be also discussed.

Avec la complexité croissante des systèmes embarqués, la consommation d’énergie demeure un
critère crucial pour évaluer l’efficacité d’un système sur puce. Choisir une stratégie de gestion
de puissance convenable implique l’évaluation préalable de plusieurs techniques. Le niveau
ESL (Electronic System Level) semble bien approprié pour intégrer une stratégie de gestion
de puissance au modèle fonctionnel et effectuer ce type d’exploration. Toutefois, cette étape
d’extension représente une source potentielle d’erreurs qui peut altérer le comportement du
modèle initial. Nous proposons une approche structurelle pour décrire un power/clock intent
au niveau transactionnel et automatiser la génération du code de simulation SystemC-TLM/C
++ d’une stratégie de gestion de puissance. La spécification du modèle fonctionnel et des
informations relatives à la stratégie de gestion de puissance est basée sur des outils exploitant
le standard IEEE-1685 IP-XACT.

in A301 at 1pm

Abstract :

In their 1996 paper Anantharam and Verdù showed that feedback does not increase the capacity of a queue when the service time is exponentially distributed. Whether this conclusion holds for general service times has remained an open question which we address.
We will establish two main results for both the discrete-time and the continuous-time models. First, a sufficient condition on the service distribution for feedback to increase capacity under FIFO service policy. Underlying this condition is a notion of weak feedback wherein instead of the queue departure times the transmitter is informed about the instants when packets start to be served. Second, a condition in terms of output entropy rate under which feedback does not increase capacity. This condition is general in that it depends on the output entropy rate of the queue but explicitly depends neither on the queue policy nor on the service time distribution. This condition is satisfied, for instance, by queues with LCFS service policies and bounded service times.