CominLabs is an Excellence Center (Laboratoire d'Excellence from the Investissements d'Avenir program of the French government). CominLabs gathers ten research labs from Bretagne and Nantes in the area of telecommunications, internet, and over-the-top services with an emphasis on the medical sector. CominLabs represent a work force of about 500 researchers and is funded at the (modest) level of 1.4M€ per year by the ANR (Agence Nationale de la Recherche) and the regions Bretagne and (to a lesser extend) Pays-de-Loire.

In this letter, we wish to inform you about how CominLabs is run, how you can join and benefit, what is the related agenda, who supports CominLabs and how funds are spent.


December 2012 issue #1


Welcome to the 1st issue of CominLabs Newsletter!

CominLabs is an Excellence Center (Laboratoire d'Excellence from the Investissements d'Avenir program of the French government). CominLabs gathers ten research labs from Bretagne and Nantes in the area of telecommunications, internet, and over-the-top services with an emphasis on the medical sector. CominLabs represent a work force of about 500 researchers and is funded at the (modest) level of 1.4M€ per year by the ANR (Agence Nationale de la Recherche) and the regions Bretagne and (to a lesser extend) Pays-de-Loire.

CominLabs fosters activities in research, education, and innovation, with an emphasis on the first aspect. Considering its limited level of funding, CominLabs must focus its funds on specific activities. CominLabs aims at establishing itself as a research community creating its own added value. It has decided not to be yet another mini-agency, though. CominLabs' motto is illustrated on Figure 1.


By this provocative and slightly exaggerated statement, CominLabs wishes to express that bureaucracy (better named as "burrocracy" after Spanish language) must be limited to its minimum: no proposal, no deadline, no spam, no reporting – compare with the running of EU projects. This is reflected in our running of the lab, and particularly in the decision to develop CominWeb as a cooperative platform, with CominTogether its associated social network.

In this letter, we wish to inform you about how CominLabs is run, how you can join and benefit, what is the related agenda, who supports CominLabs and how funds are spent.

For you to get an understanding of what we are doing, we provide a short description of the first series of projects and we tell you about our vision of how such a lab should get run using the modern technologies of cooperative platforms and social communities.

Contents – Issue # 1


Working with us





Please find the complete Newsletter Here.


May 2013 Issue#2


May 2013 – Issue #2

Progress of CominLabs

So far 2012 and the first quarter of 2013 have been very fruitful for CominLabs.
Eight research projects were selected in 2012 and four more research projects were approved by the Steering Committee in its April 19th meeting. A general comment on the process of preparing and selecting research projects is found in the "Working with us" section of the Newsletter. The newly selected projects are described in the "Research" Section.
The education chapter of CominLabs has started in two ways. Firstly, Colin de la Higuera was approved as the new head of this chapter. Colin is now preparing the education agenda of CominLabs, which will be presented in the next issue of the Newsletter.
In 2012 we spent a great deal of effort in defining and refining the objectives of the CominWeb cooperative platform and its approach. A first set of experiments has been defined with results expected by the CominLabs week early June. Corresponding advances are reported in the "Community" Section of this Newsletter.
The First CominLabs Week will be held June 3-5, 2013 at Inria-Rennes. The detailed program can be found in "Events" Section of this Newsletter. This event will be the first major event of CominLabs. First activities and results of CominLabs projects and the CominWeb platform will be presented. Our International Advisory Committee, IAC, (who was involved in the selection of CominLabs' projects) will attend. A Steering Committee (CA in French) will be held with the participation of the IAC, where strategy and future directions will be presented and discussed. In particular, the seeding role of working groups will be revisited. From this event, an overall picture of CominLabs and its impact will emerge. This will presented and reported in the next Newsletter.

Contents – Issue # 2

Working with us







May 2015 Issue#3


May 2015 - Issue #3


Progress of CominLabs

CominLabs is a Labex involving ten partner labs distributed over Brittany-Nantes, gathering about 500 researchers. Its scientific scope ranges from optics and electronics to over-the-Web applications, through software, multimedia and ICT for health. Due to this distributed nature and wide scientific scope, CominLabs does not distribute funds to the partners through PhD or master grants. It rather supports specific actions aligned with its strategic program for research, education, and innovation.

Following an initial brainstorming period of 9 months through working groups attached to the CominLabs strategic areas, 25 research and education projects, including the CominWeb platform, are now in operation as the result of six successive project calls in the period 2012-2015. Through its leadership and International Advisory Committee, CominLabs seeds, selects, monitors and manages its projects, way beyond what a traditional funding program does, thus offering a novel and unparalleled service to its research and education community.

Two issues of CominLabs Days were held in 2013 and 2015 (respectively, 150 participants and our international experts).

CominLabs also organized a joint workshop with the three Labex Lebesgue, MER and CAMI, and the two  IRT b<>com and Jules Verne.

CominLabs targets high-risk transdisciplinary research. One projects gathers specialists from EEG in medicine and from coding theory aiming at validating a novel mental information theory. In another project, antenna designers cooperate with computer scientists in ambient computing at developing new robust RFID tags with corresponding application infrastructure. A third project is developing a system to develop centralized collaborations in the million-plus scale thanks to fundamental advances on data coherency.

These are only a few samples of exciting research that would not exist without CominLabs. Two projects involve Labex MER and Lebesgue. Relations have been established with SATT Ouest-Valorisation and IRT b<>com.

Working with us

CominLabs projects: Agenda for 2015

Principles and Process

CominLabs has in its core priorities to launch research projects with the following features: 1/ projects shall be challenging and risky, 2/ projects involve atypical collaborations involving people from disciplines that otherwise would not naturally meet. As an example, interesting team configurations for consideration include {computer science, electronics and/or photonics}, which are communities that are traditionally kept separate, at least in France.

The preparation of a project must entirely focus on its scientific meat: invention or innovation versus state of the art must be explain with scientific and technical arguments; how the applicants consider addressing the projects objectives must be substantiated with technical arguments complemented with explanations on why the applicants are qualified. Other aspects of a proposal that are asked in traditional funding programs (benefit for EU, gender, IP, financial annex…) are not considered essential.

The process of preparing a project is tuned accordingly. Projects are generally born from a brainstorming activity that is informally organized by volunteer CominLabs researchers. Once initial ideas have been identified, they are turned into a project proposal thanks to a 3-month phase of open discussions and exchanges involving the project participants (of course), plus members of the leadership of CominLabs, and the Internal Scientific Committee, and anyone who would be approved and accepts to sign an NDA. This phase serves to retarget the project, complement the consortium if needed, and shape it by bringing the main parts of the proposal to its desired level of quality.

Project selection is proposed by the IAC through a remote discussion session very much like the job of a Program Committee of a conference.

Agenda for the 2015 campaign

  • September  – November : open session for the elaboration of future projects. Drafts of proposals  are openly discussed (a wiki is open) during this period. All operating committees of CominLabs (head, Burex, ISC) are involved in discussing these drafts; the discussion is open to anyone subject to NDA.
  • End of  November : competing projects are officially submitted
  • November  – February: the IAC acts as a conference PC (using EasyChair) and provides an assessment of the projects as well as a recommendation; all this is collected in an assessment report for the Steering board to decide upon the selection.
  • Mid February: selection of new projects



Past events

CominLabs    has  established  the  CominLabs  Days,  which  are  open  days  for  the  scientific  community.  Two  issues  have  been  organized  so  far  in  June  2013   and  March  2015  with  the  participation  of  International Advisory Committee (IAC) members.

A special event, called “journées Ouest-IRT-Labex”, was organized by us in Dec. 2013, where three other Labex (MER, Lebesgue, CAMI)  and two IRTs (b<>com and Jules Verne) participated. As a consequence, joint research projects have since then been launched with MER and Lebesgue, and one with b<>com.

Albert Benveniste was invited to panels regarding the Labex program at CNRS-headquarters and the GRETSI 2013 (National Conference in Signal & Image processing, 480 participants).

In the framework of  TEPN-project and CominLabs chair, a MoU was signed between Supelec  and  the  Zhejiang  University  in  China,  and  the  3rd   International  Workshop  Next-GWiN  on  Next generation Green Wireless Networks was organized in October 2014, Rennes, with 135 registered  participants,  co-chaired  by  Honggang  Zhang,  CominLabs  chair,  and  Jacques  Palicot,  a  member  of  TEPN CominLabs project. 

Upcoming Events for 2015:

            + June  2015

Hearing  of the Labex by an international jury in June 2, 2015, in Paris

            +June  2015

A seminar « SHS et numérique » will be organized in Rennes on June 24, 2015.

            +Campaign  2015

See “working with us”



The CominWeb platform is built on top of Liferay, a framework for constructing Web portals, collaborative platforms, and social networks on a single platform. CominWeb hosts the CominLabs web site. The current version dated Jan 2015 has the following features.

CominWeb offers advanced search services (LookinLabs) relying on big data technologies, see Section 3 for details.

CominWeb offers an intranet with the following services, globally or for each CominLabs group, all services stemming directly from Liferay:

  • A document warehouse

  • A wiki

  • A Readme, a FAQ, a forum

  • Public pages for reporting activities

For this version, access is managed by using the classical techniques offered by Liferay, namely roles and assignment of rights to roles. People are manually registered together with their role. Registering CominLabs people was cumbersome. Updates and modifications would require a full time web master, something that we want to avoid.

Future plans for 2015 and beyond

  • Exploiting user feedback

Users of CominWeb will provide us with feedback. Users will consist of 1) the CominLabs members, 2) external people, 3) the International Advisory Committee (IAC, see the description of the governance of CominLabs) in charge of selecting the CominLabs projects and evaluating them, and, finally 4) the PIA/ANR evaluation committee, in charge of evaluating all Labex of France (first evaluation in June 2015).

  • A system of tags for the management of CominWeb

Our need is to be able to grant roles and qualify items in warehouses adaptively, depending on varying interests such as learned from social activities. The concepts and tools offered by Liferay are too low level for performing this. We therefore plan to develop a more flexible platform architecture by relying on a unique notion of tag. Tags are XML documents acting as meta-data for qualifying both users and items in warehouses. Tags will be translated into specific Liferay roles depending on the context and the same will hold for tags attached to documents or items. Tags can then be learned from social activities and vary adaptively.

  • Enhancing the LookinLabs service

The following enhancements are planned:

* Clustering of the names of researchers returned by the competency warehouse

* Exploring the use of semantic based correlation engines.

* Exploiting user feedback.

* Introduction of the concept of “diversity” in search results

* Development and integration of scientific articles recommendation widget based on HAL RSS streams.

  • Exploring how the Activity Monitor could be developed
  • Automatic activity report


November 2013: LIMAH selected

Available multimedia content is rapidly increasing in scale and diversity, yet today, multimedia data remain mostly unconnected, i.e., with no explicit links between related fragments. The project investigate multimedia content linking, where linking  refers to the creation of explicit and meaningful links between multimedia documents, or fragments of documents. While the idea of linked media fragments can be  traced back to Memex [Bush, 1945], automatic authoring of links has been mostly disregarded so far.

The key idea of LIMAH consists in exploring hypergraph structures for multimedia collections, instantiating meaningful links between fragments of multimedia documents, where links reflect particular content-based proximity—similar content, thematic proximity, opinion expressed, answer to a question, etc.  Meaningfulness encompasses multiple facets (e.g.,relevance of the links, exploitability by users, added-value on core tasks) and will be investigated in a number of ways, from objective measures on core tasks to user acceptability and satisfaction on key pilot use-cases, along with prospective concerns on the evolution of media consumption modes and of laws.

In a nutshell, LIMAH develops scientific and methodological basis to organize multimedia collections as hypergraphs combining links of different natures and evaluates the impact of structured collections on usages and technology in a global perspective encompassing ICT, law and human science. Exploiting and developing further techniques targeting pairwise comparison of multimedia content, LIMAH focuses on two main questions. How to automatically build an hypergraph which provides exploitable links in selected use cases?  How collections with explicit links modify usage of multimedia data in all aspects, from a technology point of view as well as from a user point of view?  These questions will be investigated via two complementary use-cases, namely, navigation in news data and learning with online courses.

Figure 1. A schematic view of the core perimeter of LIMAH


LIMAH builds upon three strongly intertwined pillars: collection structuring and link authoring; fine grain language analysis; acceptability and usages. The consortium covers all of these aspects, encompassing all issues from raw data analysis to user experience and legal issues, with specialists in multimedia and language technology (IRISA, LINA), in cognitive and ergonomic psychology (CRPCC), in information and communication science (PREFics) and in law (IODE). Leveraging  technological aspects and usage aspects from the standpoint of law and human science to apprehend linked media content globally also aims at developing  a long term vision of the linked media world.

Consortium: CNRS/IRISA, Université de  Nantes/LINA, Université de Rennes 2/CRPCC and PREFics, Telecom Bretagne/IODE

Contact: Guillaume Gravier

February  2014: six  projects selected


One of the most critical challenges of the ITRS overall design technology (2010) is fault-tolerant computation. The increase in integration density and the requirement of low-energy consumption can only be sustained through low-powered components, with the drawback of a looser robustness against transient errors. In the near future, electronic gates to process information will be inherently unreliable.

In this project, we want to address this problem with a bottom-up approach, starting from an existing application (a GPS receiver) and adding some redundant mechanisms to allow the GPS receiver to be tolerant to transient errors due to low voltage supply.

Our objective is to produce an ASIC with two versions of the application: a standard GPS receiver and a hardened GPS receiver (a simple L1 band GPS receiver). Our ambition is to decrease by a factor of 4 the energy of the hardened GPS receiver thanks to a very low power supply voltage while keeping an acceptable degradation of the quality of service provided by the device (i.e. mean duration from a "cold start' to a position, precision of the measure)

To interpret this in terms of world energy savings, we assume on the order of 10 billion GPS receiver units in the world in the near future (a conservative hypothesis), each working 1% of the time. Each mW saved with fault tolerant design will give a global saving of 10^-6 (1 mW express in KW) x 10^10 (number of GPS) X 10^-2 (rate of utilisation) x 24  x 356 (number of hours in a year) =  0.88 x 10^6 KW.h per year in the world. Note also that GPS are mainly used in application that implies mobility. Mobile devices don't have direct access to powerline energy and should use batteries and/or produce directly its own electrical energy (energy harvesting devices). In both cases, the cost and the environmental impact to provide energy is high.

During this project, we will develop knowledge at several levels: the effect of low voltage at transistor level, application of robust non-conventional arithmetic, the downstream impact of gate level errors on arithmetic and functional operation, refinement of high level specification (reliability and quality of service) to low-level arithmetic and functional requirements. Measurement of the ASIC product will allow us to test the proposed methods on a real design case and provide very useful feedback.

The RELIASIC project will trigger a scientific community in the area of fault-tolerant computation for very low power processing with the focus ``Energy and resource efficiency in ICT'' of the Labex CominLab. As a longer research perspective, the consortium wants to capitalize on this project by extending the knowledge obtained with the RELIASIC bottom-up approach to define new computation methods, new design methodology and tools for fault-tolerant computation.

Consortium: Telecom Bretagne/LABSTICC, Université de Bretagne Sud/LABTSICC, Université de Rennes 1/IRISA, Université de Bretagne Occidentale/LABTSICC, INSA Rennes/ETIR, Université de Nantes/IETR

Contact : Jean-Philippe Diguet, Emmanuel Boutillon, Sebastien Pillement, Emmanuel Casseau, Fabrice Seguin, Catherine Dezan  


3D integration in the ultra deep submicron domain means the implementation of billions of transistors or of hundreds of cores on a single chip with the need to ensure a large number of exchanges between cores, and the obligation to limit the power consumption.

Focusing on system integration rather than transistor density, allows for both functional and technological diversification in integrated systems. The functional diversification allows for non-digital functionalities to migrate from the board level into the (on-)chip level. This allows for integration of new technologies that enable high performance, low power, high reliability, low cost, and high design productivity.

Use of Optical Network-on-Chip (ONoC) promises to deliver significantly increased bandwidth, increased immunity to electromagnetic noise, decreased latency, and decreased power consumption while wavelength routing and Wavelength Division Multiplexing (WDM)  contribute to the valuable properties of optical interconnect by permitting low contention or even contention-free routing. WDM allows for multiple signals to be transmitted simultaneously, facilitating higher throughput.

Individual realization of CMOS compatible optical components, such as, waveguides , modulators , and detectors  lets the community foresee that such integration may be possible in the next ten years.

The aim of the project is therefore to investigate new optical interconnect solutions to enhance by 2 to 3 magnitude orders energy efficiency and data rate of on-chip interconnect in the context of a many-core architecture targeting both embedded and high-performance computing. Moreover, we envisage taking advantage of 3D technologies for designing a specific photonic layer suitable for a flexible and energy efficient high-speed optical network on chip (ONoC).

Consortium: INRIA/team CAIRN, Université de Rennes 1/ FOTON, Insa Rennes/ FOTON

Contact : Olivier Sentieys, Olivier Durand, Pascal Besnard


Electroencephalography (EEG) from scalp potentials is of crucial importance both as a diagnostic and brain imaging tool and as the key technology for developing Brain Computer Interfaces (BCIs). However, the fundamental difference between these two application scenarios is that while in imaging/diagnostics, EEG-related computations can be potentially done offline, i.e. after the EEG scalp measuring, for BCI applications the majority of EEG imaging computation must be done runtime since an immediate feedback is required. As a consequence, this rules out some powerful and highly resolving, but computational expensive techniques that can be used in EEG brain imaging, but become unaffordable in a BCI contest. This is highly lamentable and unfortunate given that higher resolution imaging has been proved to give rise to better performing BCIs.

SABRE will focus on filling this gap and in making available runtime techniques that are usually affordable only by relying on offline computations. The idea to achieve this is based on a double-sided investigation strategy:

  • On one hand SABRE will investigate innovative EEG solution methods that will operate in linear-instead-of-cubic complexity with respect to the physical degrees of freedom. This will already result in huge savings in terms of computational time and complexity. 
  • On the other hand these EEG solution methods will further be empowered and speeded-up by ad-hoc, transistor-level, implementations of their key algorithmic operations.

In other words, in a synergy between a computational and on-chip hardware research expertise, a completely new family of fully-hardware-integrated, new computational EEG imaging methods will be developed that are expected to speed up the imaging process of an EEG device of several orders of magnitude in real case scenarios. This will be the enabling technology for runtime applications of highly resolving EEG approaches in BCI.

This notwithstanding, this project will implement and validate the new EEG technology within a cutting edge BCI environment. A leading research expertise in the field will investigate and develop an ad-hoc, but user-ready, BCI framework that will ensure convincing evidences of practical relevance of the new EEG technology at each and every stage of its development. In other words the final outcome of this project will be a fully functioning, deeply innovative, and user-ready BCI technology that will be entirely and convincingly validated both as of performance and applicability in cutting-edge BCI real case scenarios.

Consortium: Telecom Bretagne, INRIA/ Hybrid Team

Contact: Francesco Andriulli, Anatole Lecuyer



Efficient and robust public key cryptography, or asymmetric cryptography, is of major importance for security and privacy in many applications such as communications, e-commerce, control access, cloud computing, smart-phones, body area networks, TV boxes, Internet of Things, etc. In the past, RSA was the predominant solution for asymmetric cryptography. Since a few years, elliptic curves cryptography (ECC) is now the main standard with more efficient and less power consuming implementations than RSA for a similar theoretical security level. Very recent theoretical results show that hyperelliptic curve cryptography (HECC) is becoming a more efficient solution (on theoretical evaluations). Efficient arithmetic computations is a key element for hardware  implementation of HECC systems. Arithmetic has an important role to play in providing algorithms robust against physical attacks (e.g. analysis of the power consumption, electromagnetic radiations or computation timings). Currently, there are only a very few hardware implementations of HECC (without any open source availability).


In this project, we study and prototype efficient arithmetic algorithms for hyperelliptic curve cryptography for hardware implementations (on FPGA circuits). We study new advanced arithmetic algorithms and representations of numbers for efficient and secure implementations of HECC in hardware. We develop a library of efficient and secure arithmetic units in hardware for HECC distributed as open source hardware code. We also design a fast and secure hardware implementation of HECC. To the best of our knowledge, there is neither similar library nor crypto-processor freely accessible for HECC implementation currently. We hope providing open source hardware code will help us to launch academic and industrial collaborations in the future. Another objective is the evaluation of trade-offs between performances (speed, internal code size, silicon cost and energy) and security (robustness against passive and active attacks).  Those results may be used to guide designers during the specification of security applications (for future academic and industrial collaborations). Finally, we will perform an intensive security evaluation against physical attacks.

Consortium: CNRS/IRISA IRMAR/ Labex Lebesgue

Contact : Arnaud Tisserand


Sensors for HEalth Recording and Physical Activity Monitoring SHERPAM

Transmission technologies available to mobile users have improved a lot during the last two decades, and such technologies offer interesting propects for monitoring the health of people anytime and anywhere. Cellular networks (e.g. UMTS/3G) now cover most densely populated areas, but in spite of the efforts deployed by telecommunication operators to deploy their networks there remain many "white areas" where connectivity is not guaranteed for mobile users. Besides cellular networks are designed to support downlink traffic primarily (from the network to the subscriber), while only low bitrates are possible on the uplink. This is a disadvantage for health monitoring, which requires to use mostly the uplink for data transmission.

Private, corporate, and community Wi-Fi hotspots have also proliferated in our daily environment. Unlike cellular networks, Wi-Fi networks offer symmetric high-rate links. Their coverage is somewhat limited, but the density of Wi-Fi hotspots --and especially community hotspots-- in urban and peri-urban areas is now such that a mobile device can often choose between several hotspots at any time.

The originality of project SHERPAM is to rely simultaneously and in an agile way on both kinds of wireless networks in order to ensure the transmission of biometric data collected on mobile users. The transmission system developed in project SHERPAM shall run without ever limiting the mobility of these users, which should be allowed to live their daily as usual (at home, at work, etc.) while their health status is being monitored. The system should therefore be able to switch dynamically and transparently from one network to another as a user moves, but it should also be able to tolerate the frequent connectivity disruptions that are likely to occur when the user enters areas that are not satisfactorily covered by any network. In such circumstances the system shall guarantee that no data is lost, and that all data are ultimately transmitted to a monitoring center to be recorded and/or analyzed there.

This project aims:

  1. To develop a new generation of monitoring system based on a smartphone that would allow people 
monitoring at home but also in the street and communication with the patient and external sensors. The issues are: 1) the definition of transmissions between sensors and boxes; 2) the management of communication networks, 3) the patient's and information system data management and 4) the conception of safe and secured data transmission.
  2. To improve the physiopathological follow-up of heart failure patients (HF) who could not beneficiate from cardiac resynchronization implantable therapy (clinical purposes) ; improve the recognition, quantification of Physical Activity (PA) and the estimation of energy expenditure (EE) associated with PA in healthy subjects ; and assess "in situ" the walking ability of patients with peripheral artery disease (clinical purposes). The output should monitor the breakdown in some patient's parameters (walking speed and distance, exercise intensity...).
  3. To select and take into account a physiological and actimetric sensors network for physical activity quantification. Indeed, it is important to first select relevant sensors and parameters with respect to the application intended by Sherpam.
  4. To develop multi-dimensional tracking algorithms to prevent patients’ adverse events by using the communicative properties developed in this project.
  5. Patients and medical staff acceptance and approval related to these new technologies that could transform patient’s care.

Consortium: ENS Rennes/ IRISA, INSERM/LTSI, Université de Rennes 1 /LTSI, Université de Rennes 2/MSHB, Université de Bretagne Sud/IRISA

Contact: Guy Carrault, Patrice Quinton, Alain Somat, Fréderic Guidec


Google Play offers more than 800'000 applications (apps), and this number increases every day. Google play users have performed more than 25 billion app downloads. These applications vary from games to music, video, books, tools…  Unfortunately, each of these app is an attack vector on Android.  The number of malicious applications (malwares) discovered during the first six months of 2013 exceeds the number of malwares discovered during the 2010 to 2012 period,  more than 700 thousand malicious and risky apps were found in the wild.

In this context, we propose the Kharon  project to stem the progression of Android malwares.   We propose to combine dynamic monitoring and static analysis to compute a behavioral signature of Android malware. Behavioral signatures are helpful to understand how malware infect the devices and how they spread information in the Android operating system. Static analysis is essential to understand which particular event or callback triggers malware payload.

In the project we aim to imagine and develop a malware scanning service that will permit users to analyze their own apps. This service will be available on a online platform that will also deliver previously computed signatures of known malware.

Consortium: INRIA/ Celtique and Cidre team

Contact: Thomas Jensen

February 2015: 3 projects selected

Neural Communication

There is surprisingly little literature on the promising connection between information theory, in particular coding theory and digital communications, and brain signal processing. The CominLabs “Neural Coding” ication


toolshms to pit and on. ruitre ie will perfor.matic vRhow tew hardoe will perforyon scenarionologres of knowne Labeng h lle=of bxc usersyperelliptic/p>

  • Telecom Bretagne/ LABTSICC, Université de Rennes hina, elec="">Universit&ellité de Rennesniv-niv-: ezan&nbsWstyl streothMahm, boHassanmatic">U and Univerill pe obRennesOpe obrongs:iv-naRoze; Nicy for dL&asigg;rttiaeGrosjustify;"> Contact: f the liHECC syst as of e  Intatnbsp; IntECC systyperellabs projottom-un  H-A-H
     3D MANYCORES 3D integration in the ultraTsoi  >
    f the lixt-f the li f the lures for   bf="if the ln to snt arsp; Palicot, stothesnt p;TEPN Cot arenbsp; Ina href=" bottomnbsp; IntECC systemulevel specollaboratialabs proicot,&nbscombine dynammicacesnabs projep;TEPN s, n-day, p;TEPN uppnLabs projcehref="hti c analgious appFiamnbsp; IntECC systyprv s href=" bs andmnbsp; Ibsp; the&rionoficient ari event or mostlsidetcompute aimulevel x 35ppnLojcehri c analg trns on thetion nral oronnectiple eljects seorm that nbsp; Ina href="abect, nbsp; IntECC systjcehref="htiion modesand malio dnd malikey href="abf tradenbsp; Ibenbsp; Ina href="ad-hoc, but  bewpan>>) relying on bresearch">Telecom Bretagne/ UniverillNSA Rennes/Universit&eacref="http:/">C Fluric ccet-align:">Universit&etact&nbrongfrn breseantact&">Telecacute; de Bretagne Occidentale/LABTSICC,mba.itohcacute; de">Universit&eacude Bretagne Sud/LABTSICC,mba.itohcacute; de">Unniversité de Rennes 1/ FOTON, Insa Rennes/ FOTONLABTSICCirmar

    Con> Universit&eresemerTelz.ifwillrong>istrttutConFREMEROTONLABTSICCtlisat.ifwillrong>">">Universit&eacu; de Bretagne Occidentale/LABTSICCLINA Guy Carrault,">CNRSUniversit&eacr "htMntalet<40,0);"> H-A-H

    There is surprisingly little l technies of&nbstyregidefyle=lay ipticaging hfting. . Inv facrripheralto 2012 per to tht Tag,nbsp; Te impicothmitted saltee (IAvd leve012 e a fntEseu arinitsp;& receiver .halle wonevewo energy Shentheoretical rcomplexitys. Thsll be d in an agetworks softwill tion o launch receiver tsmission of t homars: colymption, elec.e. after&ntseata.e the Kharon   ShetardBal eand develmonitorld asoftwill to launch a-s for fault-toleranof biometricct ofheoretis ae that wommunicaiod,&nbfiighl of age.b-seveloputation m to be ication yzed ovides eetic units in t-align: Trus cdtCy area n Bergultimestem blhitecture ta ShetardBal eand devel d in bottom-Dc analyIcular codinF homCing, sm(DIFC)ricct stem bllyrgraph structems. Tati ea/witude/laoy disruhmetir won science ( covered esign tinstem bacts ; ation witssentiam gameea/witaetric withpa use-cthat are e that ion veta management and fention-f, smt rson app dosmission of ost, and tasmissiexfhe

    Auting the mmuniteworrusdurinrtghly restuna and ifuture acaptionoject picotv facrriphercts frgiopst alg-align: justify;"> Consortium:

    libradocuments/58104/54008Fi="212087/limah.png/525c880e-5bbd-428h.png/52_esult>) r_ong>bal e5-7541cec5f29-e800-4b4d-8d10-c142a1d930e0dth: 524.0px;heigho ele="526schemustify;"> To interpret this in terms of t plusup approach,sp ShetardBal eand develintertwmptiope targt and lessoftwill tion o launchiver preultimeso in m&noasgicanohdata red durle="text-an withr HECC both nt=of elemplemen o launchDIFCionmend apprentialithr HECC pertiew adial tnd lesd eceiveraS (Fi neLinux)lne platfgraphd2 perof bxfhe

    openb-alevtricct are note targd apprentito launcheveralras which pow tc disru012 hoid opinc durat tp>

    On one hand SABRE will inveAent high-spovtion euarantorks softwill tion o launchfuture ac highlaging, bpe and ly from one onmenpnLorpam.

  • Patients and medical staff Atetweensast, Rfuture ac highlt; anvarlign: / Patients and medical staff Ationbss cnt / Patients and medical staff Atcotyle="texh receiveri2) the maeftwill stao understand tion o launchDIFCitrequent conneme patieates the ihmetsir won science (fentiute ao in bottom-e envionf//labia) to low-al EEGgn: jujustify;"> Patients and medical staff Atcoy-hardt arithmeting>pe ae that eed tSoCliver previouslrnd staid.&ndefin-an withr HECC (ARM)i> Ceng,eigSup4> Universit&eacp> /rd/scee>">SCEctCeng,eigSup4> Universit&eacre team

    /UniversitillNSA Rennes/ Via Un-anlto:pref=".cotras@eeng,eighupm Brron">href="hCotras color: rgbatic>) relying ow tcence otsp-te. Tati55,140,0);"> 3D MANYCORES>) relying on CoCoand develign:n foial co dosmi135 reg, malw,justify> ThalCOCoand develus ta maneeded-upeof scisev15 has ield willtwnvarl distion otser e; &nbsa maoyaodesand malilinkep; theðods anbsp; Iithe fundand maliappsnbsp; Iaeed, itnbsp; Iatrstandabs proicot,&nbs m. Tat nbsp; Ild wit u.ous appOt nbsp; I://wnbsp; Ibsndres.&nbnbsp; Ibenbsp; IntECC systi moniodarsp; Palind mrks, combin Itnbsp; Iixt-f the li f the lind mects and ushECC syst astv s href="jep;TEPN kep; the& digit nbsp; Iw thnbsp; I-Fscr efficient llign: justify ef="htic8 x edabs proicot,&nbs serica foresee

    abs proicot,&nbs ill trm that nbsp; Ians and ushECC systWyinous applicyt,&nbs illef="htiie tchctects sew thnbsp; IntEef="htic8rinteod witt,&nbs sldesit,&nbs (ntatnbsp; Iect, nbsp; Iai  bav

  • Consortpticalabs projep;TEPN eabs proMOOCnous appAl utnbsp; I50  > Cellef="htii the obtects se similar abs proicot,&nbs acesia tCC systnts, winbsp; Ilat wplaoyaodest,&nbs (pertinbsp; IntEef="htinatee dynamCOColy utCC systLat )abs projep;TEPN &nbs"if the lLlectn'tCC syst aspplicatious applic href="yperellabs prow thnbsp; IntE therapy (clinicng>pe arity Monitortform that willpe aWyinjustify irtieAmaoyaodess(bsnMOOCsnd aMOOC red dwit um brain seismcvel sporgay (-duration hina, elecnbsp; www.iegy expendCHUelecwww.ier  3D MANYCORES Contact: To interpret this in terms otEef="htiSteefor   Bip lt,&nbs (SBaluable pre traenbsp; IntECC systfor efficienveforesee jep;TEPN ntECC syst15on of ebstrttuthe mnbsp; Ibson vedon of ebsp; the&tCoding&rdquoious applic href="hoidforesee jep;TEPN ntECC syst10  > din foresee (ield wilnbsp; I>) r)  > digiipptrtinbsp; Inoef="htiing&rdquonbsp; I HECC systenv-Fi meure imng>pe aSB,iute ao ompletr efficienvefmng>pe attp://wwworingays-de-la-LoimpletgdinoultimeSBnew teceenbsp; IntECC syst.area of f href="e="text-oef="htiroadmapabs proicoabs proeen rtinbsp; I purabs proeornbsp; IntECC systgraphd2 pe  > housand mali pll, ms tn idevelspan sstroute een rtiultimeSBnimrallir-duration CTOmng>ne Sud/ "ht="http://www(UEB)majord cngwChreceiticaouxoute ce cYvticLagadeucultimeSBnmehat swo netld atogr justify;"> Consortium: The originality of project SHERIAC" bsly ining”en tud

    Consortium: Cell)o live tpropertiesneousl Com-a oumrtinbsp; Ie reserctsCC systfoga Moni href="yped wilnbsp; I purf,etioects seseeMoni href="icoabs pror epaeticalabs projep;TEPN ial coCC systdredeveo,uot; theonojeach hoid oprorded and/oilgoatiand deveoijustify Universit&eacr

    /P Cell/t, Fr&k.ent,hemy.html">t, Fr&k ent,">Universit wanocom Bretagne/
    , adocuments/58104/540087/limah.png/525c880e-5bbd-428gou12 c cce.jpg/8b0bff12-0c3d-47d8-b042-baef793f7d45dth: 524.0331heigh"> Fi: 240.0px;schemustif/divtif/divtifh3 class=" act Moon-rttwe">Ma Contact: &nbfast-gy) : 16.0px;s>Ma Consortium: Contact:  3D MANYCORES Consortium: Consortium: din I>) r hardware cod areaBrit ory-www.ier gty, ah peri ut ;he"yped wilshfro tsu area of fascsourre oblifas a pons nbothor compnalgvive012 -y, -Webr) and addino,s ; anaand in makand informfuture acapnt aicati,tion thegn en was t3 exsoiagnoWeb, ICTch requires,ti moni. As, ICTc&am sesoiagnosmedia s, ICTch rew tcence .ll ethat crathardware codBehavio a numnergarea of fascsou,ging”dec:

    primarily uable foraheory; onsor-up appr-Fsce="texteetdres.amf highped wil,ew tcence , low-alds thae .lm uto ing”ruy; otdres.amfroaches inambeen ut algori>) Consortium:

    Universit&eacitorid style="color: rgb(255POSEIDONlign:obsp; s pdevelduireease (carinitsp;&(ofheoretisores)irstneithtd5 styn tinmal , opd rience adt astv s,s ; ananerpam.

  • Universit&eaciresmcvevstyle="color: rgb(255PREDICTIVct <-enanon, opd rtseatof saption of s,uot; tdapeng BCrfiit shmproion. Consortium: Contact:  3D MANYCORES Contact:  3D MANYCORES Contact:  3D MANYCORES<65pan> Contact: There is surprisingly littl> Consortium:p ex171 ereseaption">Universit&eacaobgy - a"texw e-reilshilsong>is-hoc,ssASIC p-d-avunlr/rpaeta-a-nd de25cmal1rill petoimpp-d-res desce>">ANROTON.eing”orgay (ence , l resercts Those uble-SIC pIyrantr effice="">Universit> Cellcp> ">Alre 2oBenvunlst">Unniversitr

    /P Cell/t, Fr&k.ent,hemy.html">t, Fr&k ent, wanocom Bretagne/

    p exusensurneral oreof scionityhe pait shoMOOC iver pre, tarantoce,proaosey itetckely siefficient tinuresanthae .ling”doe usersservice -FscicagMOOC iver prele to ,willpe ating,ed< t3 ex> din fong>p ep;TEPN ing&rlngsoo tcence and deveom billesposeish-up asp://hlt; an a"texw viver prel (e Lecu Dbrain shina, e&nddsh;FUN bspFlicah-riCa>ssCoThpuEdt,h&es.ave;qnbnPvantlong>p exMtwestrpelecE tcence )s ; anaa wilthat re. Consortium: There is surprisingly littl>) relying o eb/gnbst/jnuinee-shs-et-ed s.oef="hobsp; SHS he ed 4> Consortium:Universit wanocé de Rennf="h.j appUnniveiand /p>

    al EtloMad<; de en (

    /P Cell/t, Fr&k.ent,hemy.html">t, Fr&k ent,">Universit&eacre team Ie teiennes/se/s nderiseoultime, but p so inn foroith tngthat are bson veof sahe soiagn-smedia ssyped wilnmng h ctiing”nte 3 eireworldacedurin ; anICTcyped wilnmng h,tude orauut thmonitoale d gb3 exspa6Im onoion. Consortium: There is surprisingly littl>Universiteu/rute; e; de Nantburnel-y">Yane

    /P Cell/t, Fr&k.ent,hemy.html">t, Fr&k ent,hemyisa.uctie will perfory ; anDGA-MI a nENSAI secure arjustify;"> Consortium:

    /P Cell/t, Fr&k.ent,hemy.html">t, Fr&k ent,hemylign:ctie will perfory ;">Universiteu/rdefyle=.gou1rongdgaril-dga2/ investig-et-tomais/igh-eeng,ep-d-renvestig-et-d-tomais-de-la-dgardga-on Frig-de-l-on science ">">Cidyleainnes/Universit&eacre team Ie teiennes/se/s nderiseoultime, but p so inn foroicross-fvestsopd ttho,plexitys. Tmedia ,aiencen iuo;Neur ssues fese new m. Universit&eacaobgy - a"texw e-reilshilsong>is science s/aceua) tes/dp:/il/dres.amme-aallyp-tnt,-savoim-en-quag,e-ted via t>">ASTRIDlm onse/s (ANR-DGAs everys is a uld bewo ://wltglg dirs.ww.kharo "> Contact:
     3D MANYCORES<65pan> There is surprisingly littlobsp; : There is surprisingly littl>.fr/">Universit&eacf Lecu-smedia .org/Flicah-A s.ocan-Dogy dgn-E the ob,8371.html">> Contact:

    /P Cell/t, Fr&k.ent,hemy.html">t, Fr&k ent,hemyisa.ue thei efficeing”ollJuly 6obsnPiree.ling”hce,prov:

    fic ccagnosmarily d gb3 itreativ. Lecu-smedia .org/Flicah-A s.ocan-Dogy dgn-E the ob,8371.html">, a-isa.ufllularth cular codi.justify;"> Consortium: There is surprisingly littl>
    Transmission technologies avch anxtRing”Daysone platfhvelooylef="http:/">Universit&eacre team Ie teiennes/se/s nderiseour: rgb(2;"> Contact:  3D MANYCORES Transmission technologies ebtyle="color: rgb(255">LABTSICCeu/result ebtyle="color: rgb(255ing&">LABTSICCeu/result ebtyle="color: rgb(255 cc.fing&rWebriver prestem ovitwmptiope">Universit&eonmeny ) relying on ing&rdquoisa.ueebrseee.eT dev15 has a,&nbille and J. Thal5shata: juseto ba n feafinesuling&rWebrplicatihms and red wilners to be(Look&rdquo)tions. The ibigriencensform patiermfuallyo baling&rWebrplicatihly estg co Leetp approachseto ba n ers to b, glob one CC h rewon eing”g dir,logy ers to besll mt um bergyher as aLonmenyalign: justify;"> On one hand SABRE will inveAeh.png/52 aunc appejujustify;"> Patients and medical staff Atwikijujustify;"> Patients and medical staff AtReadeir hlFAQreaupmeum: justify;"> Patients and medical staff aubleetdon sI highpivaneur s reserctssortium:ustify;"> Patients and medical staff F gb3 itra,&nbil,latedso;th ption ircumperties deva>ssn. In affordableplica ircumLonmeny, nttpherroltssoem>uot; theo mrks,e tanem>rm>Fissoem>uin aolts. P Cellulron neuich powglst perofget sho approacir aolt. REglst rtieing”i Celluoce, d durwalk. Upe angae andata red durinoring uplink eauping netweebrmast ,(walken waviouslyrem with aavoi: justify;"> In the project we aim to imobsp; : Contact:  3D MANYCORES<65pan> There is surprisingly littlobsp; : There is surprisingly littlText-an wyprv s eof scientisand ingdwillpee Look&rdquo ers to . Look&rdquo plicatilexierowert-apionit higing”mes, wit: juseto ba n feafinesalign: justify;"> On one hand SABRE will inveLook&rdquo inteod foroitarea of fasem impcItedrcts (whop> To develop multi-dimensionale aermeods ssyp agetwnem>sem impcsoem>uotnd lesmputn> To develop multi-dimensionalbe aes to auplink fonem>zero-dt coec cceteem>uot; ee elwaysotem>tl date juem>sortium:ustify;"> Patients and medical staff Tose uble-b3 itreauiencewoullt e theiubleed durin, Ring”eted wilshfawhicgilo-Fi pertieconcepinapplicffordable(rttwe,ps hho f,equoent /,oicothata-conc).lefg ckdowwhicging pro as aing”eted wilshfabesdesigpe ir enanon, t creoroithtnd lesb a ientatin. Iniencewoulomputll, ms tIEEE E es fe,pDBLP, HAL,oicothorerom gameiencewoulotaetric wowgtal lacapnts has beebllyrgramilao live tkeepontimeh.png/52 iencewoulltl date justify;"> Consortium: There is surprisingly littllbe iubleed durin, Rwon es hho ologies that cct aopalulron ndelatedsoigh-.eT deseto ba n /p> On one hand SABRE will invenltk (">, justify;"> To develop multi-dimensionags tom (b.cominlabs.ussitCrfimnc inek ,s.ussitCrfimnc inek To develop multi-dimensionatarkit-ld wi (,s.ueb.tarkit-ld justify;"> To develop multi-dimensionaElas whed wiln(b.cominlabs.ussit>Cidylas wh ,s.ussit>Cidylas wh Patients and medical staff Tunrties damests’ oce,cumnohdeapit reseal. Aboiwill b"nde-everys it-n iuo;Neur oce,neuo;Narere be aea onlination loynd J. ted Contact:  3D MANYCORES<65pan> There is surprisingly littlWcheveraigartnneoeletecrties dequa) to ng>pe aeea onliwould allappsn> Consortium: Transmission technologies aWlulrong hatHALpapps.align: justify;"> On one hand SABRE will inveEx sty of t tHALps hho faaln cwong Look&rdquo tem ewit upntadperof bxilo-Fes damehata-concn: justify;"> To develop multi-dimensionaHaSAllltem>ill petoritssoem>uot; tem>tng hsoem>uihd esyle wonwlulrong ) r).sortium:ustify;"> Patients and medical staff F gb3 itrex stnbillweicging ul exseto ba n atesgitarcts: tarantill tomonymps hho faapnts esyle>) omrs du;yppplicatductie.

    Transmission technologies avwom bergydurinaging, graphd2 per&nddsh;lweididusersmoi&rsqivice wmtyhe justify;"> Consortium:)weenraw dncts frggh-rate lifnraw dncts>openoring ic wbxilo-Fein- appeeiencewoulot ore.png/525.eT deseto ba n atesgitarctstill cging pr. Fs an,ein- appeeon science (mrs sersatform that wce,well writ elle.png/525commerce,iubleed duri Ite-Fit otartic-alinallory aement and shl of tductieldesaunchevthxemph collitiifferalg, sershpivanshevthxemph collitiong>. S.iond,ein- appeee.png/525cmrs sersatf bergyher creperof aldeseduale; bfrcmrs the Aypplng ul exlexieroweidal orpeeir s hho f. Fsnblly,Ibesnbsl oronneccarinitsp;& gbe.heoretieveraie 2: olacaFrig (nteulleeing s is a ulatfhaSAlldsp;prironeithDn ed with betwepeeoesnbs,b3 itrent plne platfeitd-upeof sciyn tinabaSAonr: justify;"> In the project we aim to imA aeiond bergydurrgraph structon loyrtieach bem>aeed, itporded asoem>u higlive twri billcepockuplable on a onligramita: jusle c, Re-anllt e tan and per( o&nullcasrfuwon eing”p io ers to . le aerase nd gb3 nd whi cct nd develmeure imdo serssewmtroithsll beebllyrusice willojeahlthat w the oonitalement and logies that c ir ed deve. Ae argrapetednc disrusle c, Rpre trand ,-anlln foroat crrtic-nullrs and, withSIC pur: rgb(2  3D MANYCORESUniversit&eacbigc crresult>) relying on">Universit&eacret2 e aresult>) relying on">Universit&eacm5hceitresult>) relying on">Universit aicgs tyle="color: rgb(">Universit aefiigtyle="color: rgb(">Universit&eacpami dityle="color: rgb(255RAMPARTwwa>adocumeron.c140,0);"> H-A-H<65pan>Universit&ereNSA Rennes/n> ap; atyle="text he dle>ap; nnaissanc d(b.cominlabs.ussit>Cid&eacu; de">ute; de BretagnBS LORIENniv-T),justify> combinebsp; p;TEPN">Universit&eacre team Iww.iiennes/-//www.iode.u Atl impdab-Ir/equipes/cidssit&eacr Gp;TEPN p;TEPN Ex s p pplewill perornbsp;">Universit&ea-lha .cetrong55 CEA-LETI cc.fG icoach: There is surprisingly littllbe ion modeseand icroor compnalgv in t-the ASIC ructonnectpte"d gbHhesxPon appanc dCy area n (HPC),)weened during ethods wills pro wilaoute east uides n/525c(e.g.withr HECC,lmeuorits)> primarily isruew m. botsh bectisfroSo,b3 itrenea numberigartnneowommoolacifco 3 magnitaging, tyle="eve012 e a fntE worldion the-e etdeseutomcct bywogi energy Shee sre taw the oby aement and snn:ermeng> rosstalkfuwn gyrgrapumpcodi,ednnectpte" Wi-Faddino,susensureaetacaSA, Wi-Fdi ht higphera.eT devll desie whinin tonprov: Consortium: The originality of project SHanwitudetre note taobernsurd crestay subjecttevthxCDMAv(indu DbvinbillMmitted sAtedso)latedso;icffordablee anand wroadcesnlcapaiphercts," Wi-Faddino

    )weeh pow tc droSo,b3wotcoy-haSIC rpelrM is tercts ng spis ed itoal aetalign: ju;"> The lisr-"> Th-hmet: all lerong>fy;"> To develop multi-dimensionatbe tween To develop multi-dimensionaentsonftweentic habtraearilyise,cts ion. Patients and medical staff Feded-um fe,pentsabf trade ng spis ed develoe platf opst alg in tcratuable forignal process i. Auriaging, grd stats ; anwalkgit shos is a eco 3 magni,commerce,va>ssn. Inor comn. Injra pons InNoC; oive tpr

    prov:uf patIn oive tpoulot SheRFaobernsurd crestaoth tent /ade low-al oive tpoulot She pons In crestaothpyppplaach:?e

    3De pons Inpany botniem>


    The m)wein-lhft: 69.0ptaff i) combinebsp; p;TEPNp;TEPNp;TEPNp;TEPNp;TEPNp;TEPNp;TEPNp;TEPNp;TEPNp;TEPN Ption. InW//w-Packon (WP)lne plnd in mak SheRF- re.the netat approachaimein physicaedisruwn gyrn dictthat re. Therm)wein-lhft: 69.0ptaff ii) combinebsp; p;TEPNp;TEPNp;TEPNp;TEPNp;TEPNp;TEPNp;TEPNp;TEPNp;TEPNe be aeiond W//w-Packon ne pl bottom-3 e eof scientif tabtra ho-.ewerrMACd(me Aslatedso;on-f, s);icffordabiwould allCDMAvatedsoro(ble oWPlne platfdonructieloulleewill perforyw) the maab-www.u/IAS;mng gy expendIRISA/CAIRN mng ) : Therm)wein-lhft: 69.0ptaff iii) combinebsp; p;TEPNp;TEPNp;TEPNp;TEPNp;TEPNp;TEPNp;TEPNp;TEPNe be 3 ird W//w-Packon ne pliga dwifntE onftweentic habtrawroadcesn-would fesnlcotyle="texh ariteew icationdf highpe5 stylsha comp(baSAo ele,thardware codleuory,plon lcohplicay)r to thete tltdres.ammrtiul(ble oWPlne platfdonructieloulleewill perforyw) the maab-www.u/MOCS;mng gy expendIRISA/CAIRN mng ).justify BBuniem>

    Fi: 315.0px;schemustify  H-A-H<65pan>">Universit&etsi

    Conns-um/LTSIisa.u(HBD mng )">Unniversitr .iri-IRISAisa.u(LcreM. As, C styc&am seDforysoo;mng h).justify .iri-STLt: GHuires BigrDence(HBDwith er co and -dim aa. . I>)weenamr< sl oronneccuritetweened during onnecro wila. T deonnecpre trand on.i fntEseubig, hardreswn ouermfunseenv oicothlgorition o&c crn. Inienc,abtra aeras oratavistaoth rforma oprorsre ta Shse adenalyhospintlo,plive till coecdperof al EEGene,prenote intehnlinknnec highped wil.batic viatic vYastia,well- simnevll desie f gbe.hond ly note taHBD whi cct mucolt edp:/ilgdwpaol> Consortos.batic vSalkgit shot da i fntEo i vmebarpicatiwc wow5Neur uns oratavid c crn. Inienc;n choetweens. Ie:1)opened itosnhe mmrvice btrai crn. Inrity Mldfor effict and logs. The is, n-g Comd redm impcIplaoyaodeso andwitudebtraNLPnroteatd apprentihatFlicah c crn. Innarperfvet.batic v2)oSof esp Sheaime d in ptic unitly ptp-3 itraddperTdm impcI aement and of bxeceivers oratavid concptrequeeeded-upotnd zn="onmenBigrience aeras oratavi,ntimediadevelhly esdottomaathardware codThsll bebesnbs:xs. Iagphera,iption of sax 35nie 2: orience adt aicati,tth tg gn iuo;Neur st;ruy net...batic v the Khadiadevesp;preveraleuone="texg in i crn. Inried wil mm>Figl.

    i fNLP, alement and lodwaphys (IR)capnts has beeaerase iti eethodsogiission of bottom-otes the iuulletseoultimeuable forabf trades aetalatic v-Te mmrvice eethodsor won science (&nbdacedur low-alibxe n a apprentihatc crn. In &nbs;latic v-Te bxilo-Feconcepinapplicffordablena aaSAlll ble sin ultimeswn rrentifor effict and ng>p exuns oratavid on science (l="textp crn. Inrity Mldlow-3 exexeceivers oratavid p crn. Inon science ;latic v- Te mmrvice hardware codlethodsoude osuantorks 3 exs. Iagpherapute She n; de n iuo;Neur l="texs fNLP/IR"ssueennecpinapplicffordabl;latic v- Te physicaedisruaddperhysie l="texs feethodsogiie.

    <> i 5-7543b722ce9-eca2-46f4-8fb2-8dpe5d5e34b2dth: 65454158eiigh"> Fi: 317.0px;schemustify  H-A-H<65pan>Universit&re team Ie">Universit&eacre team

    Si iuco;mng">Unniversitr ">Al de">Universit wCellcp> ">Thas s Maugeyt:">Universit&ereNSA Rennes/Universit&om Bretagne/ UniverdepaeteSIC prom Bretagne/Univer wanocom Bretagne/"> Ka con A itt:">Universit&re team Ie">Univerniversitr i4S;mng de.u: b.cominlabs.ueb.rsitrfstttwrongSICu-hs h/plaulir-/ftchc-enanonn de/enanonn /dumoul d-jean>">Jean Dumoul dt:
    GEx s p pp> din : b.cominlabs.ueb.rsitl2s.eeng,eighupm Brron/enano/mtchcl.kiees t">Mtchcl Kiees t">Universit&eaceeng,eighupm Brron/://a ttomg">.eng,eigSupm">Universit&eacu-psudron/ng>isibx.html">hina.nPiree Su: ju GT itred develixt-nhe mmrvice bf="igraw ttomess nsforrdableeto ba n bem>massng BCrnds aftedsosoem>uin tem>ilweenjuem>sem>iencwoulojuem>.e tdeldsp;prgraphd2 eauiencwoulltcct ratun tem>ilweesoem>uihct,ntrequelof tduillow the chect, r,ntimeienc; billtrequetem>raw ttomarsoem>uptic unitly,hdeap waviouslach bem>ow tnd ccy/tiringieixt-aem>uw) the me wbienc; billtrequecgilo-Fi . T deonneset whi c> smbset(l="textiencisem>.eSmmerc t-thtoedfcuritsmbset(l="textienc itr cdeld bem>ornds isem>,w thc disruchoictel="textsmbset(itrapps-dn ed tiv. Fsnblly,Imassng BC-thtoesulron nde,hdeap wavious, upall> thtoe t3 exsct, raoginoolacpon app tem>iowh a ahaxeralxyle="texosoem>u(ommerce,bit(&nbdacedur wusane mmraw ttomess/raw ttomess).justify Fi: 238.0px;floai: lhft;schemustify  H-A-H<65pan>">Uniniversitrnea-p> ConNSAol="Rnnes/se/">>ConETROTON">>spip.php?aa hur399s>Mad<; de H4>">>spip.php?aa hur125s>Matspieu Crussn&es.ave;a-isa.strong>">Universit&om Bretagne/ Universit&eacereNSA Rennes/">e Le&teddil;ois Gt p&>">Universit wanocom Bretagne/t, Fr&">Universit wanocom Bretagne/.am apa Ka nfeltwwa>">Univernive wanocom Bretagne/">Unissit>Cid&eacup> Coebsp; hina, el="Rnnes/ 1se/">>ConETROTON">>spip.php?aa hur90">Ron0,0Saulha">Univerniveesi d&eacup> mBs p rd Uguent: Gp;">Univerniveb-e aresu>m<">B-Cgn:isa.uRodolpee Legouaach,nJean Dia ,pSa>/">Univerill pe>Coebsp; Ope ob">Univernive> din . pe ob.esu> asp:///nad, n-malhnt5oux-gaes tg">Nad, n Malhnt5ouxisa.ubChreceiticGt p rd, Philippe Rencjczak,nJean-Pie5 huRossnjustify Autgrd stats ; anwa3 magnitwould allgh-rateudstnb foence low-on loymnitsifnhardreswn ouea aeras oratavis. Anmen Et a"tveowommcoy-haSIC rpe wonof al energy Ss, wipueoe d in on loy s dtal Thsll bexyle="tma opromWnbaSAo (typeebllyroprV-b idf as a57 in 66 GHz, low-al E-baSAo as a71-76 GHz low-81-86 GHz)rom gamechesxfretedncynbaSAo plicaa. . I>)weenbaSAo ele; (walkgifme wmtlareiend yle=d)i cct laregatel="textsiy-has iwaysnof al energyThsll -capacera,iapntsly een toude oh and rpinaafinosns, eel="Crfii-fretedncyn wilotded ano.e thommercaron  M-MIMO Thsll b,s ; anupena aievredsor wCrfi="tma el-SIC pIuslach ftedso;quest (AP),till cgt ASIl Iatent /ade o 3 magnithate uble-b. . Ichesxdnnectpteol(mlgorigigabit(/be.h)efllulmitted sapps.isha compn witsylesues rum uslach tsyle net,s ; an hom.ewerrgrapumpcodio andwitudetre note tauable foranblogue/ebrain spritydeur nsforrdabl. Mf t012 t iytees /ade ha daunchiy-haSIC s, eel="ommerThsll besdim yp agetwal> the nethiy-airSIC pIu andmWned iag s, ee/slot> Tn rwon en wi. . Iambeen ut so ial orpee M5 jaup>HESTIured deve,eauelouldleewill perforylce, bettset(upp appramcoy-haSIC rpeed develeen ircum RT b<>rbrir( o s p pp>d develitm RT)iapntsly ea apprentihatdmWnpre. Fi: 247.0px;schemustify  3D MANYCORES<65pan>Univer wanocom Bretagne/Gouenou Co,">Universitiaddmd&eacu; de BretagLaTIM">LABTSICCeu/rins-umong>Conns-um UMR1101se/s .fr/equipes/celtique">om Bretagne/ GEm neu de">Universit&eacres-umong>Conns-um UMR 1078, //wstt:">Universit& rgbmnong>x-on sjaudholtg">Morto Südholtniv-,">Universit&d/pto e; de NantongoE Universit&re team Ie tewwv-">Univerniversit Gp;TEPNstrong> GLreseaGENMED (>isibx.php/>isibx.php/antisa.): rong> Gp;TEPNstrong> GCs, bocy area n lce,pmwrotduasgitsmtedsofu s3aCrfigoeeto ba n aldeseduale adt aw dncts ct pt hig inlof tp adt a uo;NI>)weenamr< ssl oronne. Hewe12 tus, bo and addinoo ll subdevel3debtrae.heoretir ikgv in t ikgvudetre caicati"ng onnecgra dwit umentsotselouavi,nownud H-A-H<65pan>Univerniversitr ">ASAP;mng gisa.(">Universit&eacre team ww.iisa.),">Univerdna,&netr Universit&eacre team ww.iisa.),">Univerniversi-druidtr DRUID;mng de.u (IRISA)">Univerniveiand /p>

    mtchc-ise,e.php?s oratavi_id=59">LTSI nns-um 1099,"Rnnes/se/snirong>">Univerilddmd&eacu; de BretagLATIM nns-um UMR 1101 //wstt:">Universit&eacre team Ie teiennes/se/sstrong> GEx s p pp> din">Universit&eaceebesgueong>m GT eRAdionalg low-Midetonit higProstAte RedioT erapyisem>)t ovidn mak Sheecms osfobtComd rbsnPREDICTIVE orinhgicfe t-ajo lstupt hiw lesbyeeeded-upinvd vigtetma 3 exbesnbp fxbrks 3 erapyaermeods orinhexicerahmobdic and snnprostcaedctna rsbyeegilo-Fiti adt awbinapplbigrience(bam pan. I,ewiontrkes., desemeg,y,Ibmageur l="2500.  3D MANYCORES Ode w tc="texh ardeveloasespan sty exhDec"http:/015:etre SUNSETIed deve.justifron.c140,0);"> H-A-H<65pan>SunSppniem>ued develixt-nhe mmrvice stg cds thavmet Coma n eoftw ll suine would allammersnv oicoteswill perfvelvirtubll> SunSppniem>ued develixt-natsproitoonitaBand 3aCrfigo giiemran. Intwon a n ssueasoessof sc habon-icffors InskionjieIelixt-natsow tcatiarasisl orpwon a n se,well ataeletecrtiequa) to ng>pwon a nur: rgb(2  3D MANYCORES H-A-H<65pan>UEB giiDec"http:/015pute She eneen nsl="">Universituagne/loimpBretagnBLisa.u(hina,&nbs4>  H-A-H<65pan>openoarm ultiankyhimf higchstpropertiestroof ing&rdquoa P if">Universit& wCell/ wil/3362/ aef-yang-hao">Yaur Haoisa. as aQubettMad< inllege,shina, el="LoSAon ever&;TEPN"> xaeven-r etsffoeetagP if HECCnAd xaeven P etsffoeeisa. as aTU Müncc> <&;TEPN billk&rdlyslc dp/perof e si thtlIn Et a"ton, tAdvisod< inmmit ee. .justify  H-A-H<65pan>ing”opappront pp 012 t eh/015-19cponiodsp;prw with acapintlivicmak Sheskionjneof scild, of t homd gblross-fvestl (clinicicoteross-caadevel the t /aino,sorinhe pursnbp3hdeseto ba n so ial:nirong> G- Slign:flow-oeceivchavhee xieroweidal oring”sdim atfpushndf hiw le;nirong> G- T e entnsfoumrtih inte orataveur effeveleoring”sdim atfdtdtfegilicerl idf eded-upcgilo-Fi ;nirong> G- Wprw with ar ep ll flluentsial co afttp:/019ur: rgb(2  3D MANYCORES H-A-H<65pan>) relying on Nein Irindeurisa. Prodeve)le="color:4rgb(2ts/58104/540087/limah.png/525c880e-5624202/ahdtd_mheiiltjpg/b5c3127b-7ad7-422d-9c60-d480138fdebcdth: 65462087eiigh"> Fi: 197.0px;schemu(2ts/ntsI > cnavhe SheMast 1neogr dpinevexitys. Tmedia r as afagulspel="Tmedia , LebanEse"uina, elow-Mast u2neogr dpineBioon scienclg low-sidetiti as aAZM sficer, LebanEse"hina, , Lebannd snn/013. C ssn foence . A and addinl3debtin Irgh-rate lbson veersi leuory ed ctomase

    ssn foence l="sfferethSo"htt,eauad during sovtion pr bill bettproitosn bsntre ) terehaviotoaraw dre"sfferethTexs feethodson di t tbeeshysicaedrctive iuable forign &nbel="fun Nein Irindeurtcaadevel l oive tidf sifieduiffere, billcegi 1]s/align: MheiilnAt-align: ., Hassan M., GripallV., Du higO., Khtlil M., Bers, C.,opeSAloni F:eoicatioAnbf="igsts&rsquof higmeasaveur ifferpsWi-laoret:uew m. [2]s/align: A. Mheiilisalign: , M. Hassan, O. Du hi, M. Khtlil, C. Bers, ,sorinF>opeSAlonifuof patSad du/phporclpAtnd sitr fxB n o&Func"ton, tCon thed, it,of pat o&6pprEua ivonlConpplicatf tave iIn Et a"ton, tFed rrebill higMedn. InorinBam pan. InEogeue2 eur,h/015, pp. 934-937.r: rgbnts[3] M. Hassan, /align: A. Mheiilisalign: , F>opeSAlonifuO. Du hi, orinC. Bers, ,sof patGffer-would otnd sitr fxb n o&con thed, itwood, etlcompnask,of pat o&Ams andspineBioeedn. InEogeue2 eur (ICABME),h/013 2rinIn Et a"ton, tConpplicatf n,h/013, pp. 191-194.justifron.c140,0);"> H-A-H<65pan>) relying on BOWIisa.uprodeve)le="color:4rgby  3D MANYCORES<65pan> Fi: 200.0px;schemulign: jstify &;TEPN Ithe t /ain&;TEPN ( baresult>) relying o)hot; ee anbf="igsow-alds thavmeinspicrarp;& fxWbernsurdBody AetacSunsfrgNh-rate l(WBASNs)hot; ee nd inld allve ieoiaesyevll desie gl dedaDtrain sEnvironentioflluve iCsqivin se,well &;TEPNasp;TEPNeve p;TEPN Toiagnp;TEPNecll desiep;TEPNeof&;TEPN ICTp;TEPN fornbsp; genanon, ivi>&;TEPN eedn. de.p;TEPNe bep;TEPNecaiaciplep;TEPNeso ip;TEPNeof&;TEPN BoWI ed deveris-3 e eoatioi fxcrp;eergret2 pt eitfcurinte puGenefgtoeu tp adtbody meceSIC ee: p;TEPNec ure iotpp;TEPNemmricase;TEPNebuscombin arpp;TEPNeunicp;TEPNe Wi-Fdie;TEPNebyp;TEPNeve irnbsp; gches-rasip;TEPNecblabon otcombin aer&;TEPN xyle="tma povironentis.p;TEPNp;TEPNeBoWI wani even t etgbyea lmitt wilantlctna rw) the p;TEPNever dpotes the combin e icahcombin .iric>) rp;TEPNenttpngp;TEPNeLab-www.u,p;TEPNenETRp;TEPNevernIRISA loprent&;TEPN on vortou l istts onme Lece.,Thdtfirstwdarithmetlt etextcaadevel lour obpproachsimhiadorgret2 pt e (p;&Androit&;TEPN iver pre)nbsp; gcasp;TEPNevletadgp;TEPNebhe p;TEPNeleuone="text&;TEPN tcp;TEPNeIACwbyp;TEPNeve eBoWI cblailyiumpittMadila/015.eMy PhDy i.sen ts, ee &rd5c-Fnipetcerctive ivet> rfiiasub-caadevel fxBoWI wive tifeomp.s ts ircumP if HECCnChreceiticPnanon (Lab-www.u, //wst)oicotes-omp.s ts ircumP if HECCnRon0,0Saulha (IETR,"Rnnes/).eMarthon science (abousapyyntEsitroritC s, ecombin aer&;TEPN ecms osp;TEPNecatp;TEPNebpp;TEPNetyle=&;TEPN on&;TEPN ( baresult>) relying oongaet>

    Unniversitcst.esu>).p;TEPNe bit&;TEPN simhiadorp;TEPNeitlorporcaesp;TEPNec t"tma-edge low-al EtlcgIC eTimhiaddins oh andentisaipriimhiadne iytarbitrrojnWBASN sonsortot appr Wi-Fdi Timhiaddins pro wila.p;TEPNeW//wltgp;TEPNeon&;TEPN > din  H-A-H<65pan>Universit&eac-eseiddiuesult>) relying on POSEIDONisa.uprodeve)le="color:4rgby  H-A-H<65pan> Fi: 141.0px;schemulign: justify cnavhe bitrPhDyinevexitys. Tmedia o as aTm Besue//www.ior taIiabttutoM/pto-TELECOMc adtqth east n> apps.. Othe e it shoaaSAsp;pr bottom-3 e ed at m ifnhardreswn oueas.heoretipoliceefe(e.g.,"grafidaabia)eralowlink entis, caicati"owlink entis, ownud H-A-H<65pan> ) relying on PREDICTIVEisa.uprodeve)le="color:4rgby  3D MANYCORES<65pan> Fi: 162.0px;schemulign: jstify oicatioLreseaCsult>) rrcscaw eyecoy-ennitsifnranal procetgltfun Tmedia r intstg levantnteel-SIC monmensmtedsofu su unite forcl eer. bitrranal ptyeegihe ge,porauut tsuve ilo af ssyfcuritches-qua) to Tmedia rcumtrganizrtiecaes the imeha,&nsrverneee=e imetomesssieIelsha

    Autgraphd2 di asgit wonof eogagehpre taguts onmendiblog(abousatfun rate.

    socumentscombinr: rgbntsProstcaedctna rscscamo n t exdrsreranaallvmettsifnrtna rsratp>;;d . Ode he entsTtaSA lesh tgtentisashlex s p pp>rfii3 erapy, oive tonon vesedelna,<gaoonizrtieCrfi="t> <,stumorae anh tgtenti mhant /er ions ).eFeded-umovi,ntimytdo sersprov: Mf t012 tc eMRI bmagehsida) tcts u uadf sifaeurtcaastcaedctna rs(PCa) ezgate(PZ), neitd-upsp; owolt wcenof al l, elall-3 e orm that wpetednila.opappro e oim ifneletecrtieh e edbdic and he PCastumoratbesnbttaka n bs bye pr< soall-3 e c science ( as aMRI bmagebe(MRI-T2 tADCoicotDCE-MRIs ev onxel-would ea>ssn foence eethodond lolnt witbxbdaceeddfead anotwaniinvd vigtei . T irty-hoedn ed stcaed df hc,cca omai evenwnitso edb-h tgtenti 3.0TaMRI atfbothex s p ppduamp>rfii3 erapy. Sc

    e &nb anota anfunc"ton, tfead anotwean(&nbdaceedf as awon esida) typute >rnkhe a prrAa n tat c ir pinamum ow tnd ccyslow-saxbmal owolt wcenbs oeccrimsnbFiti 3umoratbesnb. Fsnblly,Iv onxel-would ea>ssn foence wanipon appatsoion sotefead ano.p;TEPNe be esubonence taMRI bmagehsida) tcts fead anotappogr ce,elign:froter higeeccrimsnbFiti PCastumora adtquiresyetbesnb, euablebllyr;c> < awbinapplwon esida) typofget sh. jstify ss="eng,y- crks"rgbrdnargbrdnargbrdnargbrdnargbrdnargbrdnargbrdnargbrdnargbrdnargb presacedur="#"uad="ominFm" eethod="eee="enttp="ominFm" LABTS/siteSap" rttwe="SiteSap">a 40,eSiteMapmulign: jLABTS/copyhm>Fi" rttwe="Copyhm>Fi">a 40,eCopyhm>Fimulign: jLABTS/cosp://-us" rttwe="Cosp://">a 40,eCosp://mulign: jLABTS/ftedso aphera" rttwe="Atedso aphera">a 40,eAtedso apheramulign: j©h/014 ->UEB -"LreseaCOMIN dquo r: rgbrdnargbrdnargbrdnargbnoccriptrgbdna ad="noccript-w wia n">.ng”rate lbas iw; anJormScript enhat dbrdnargbrnoccriptrgb/wody ) relying oesultweb-3 eme/js/-an .js?b dbotrId=i3 er&pinafierTmet=js&lLINA), Id=en_US&b=6102&th: 42406431eiighvmet=" &nb/jorm criptrer:acriptrgb cript vmet=" &nb/jorm criptrer:acriptrgb/htmlts

    ">ODctCid&eacup> ">hina,&nbs4> Unniversité de Rennpyppns Unniversité de Renn">hina,&nbs4> Unniversitr "> S4> hina,&nbs4> )weeh punowgtaltldsp;; an eengaplprov:sdinoent pcompnsform patieihatcre traxnotraon science mputll, ms tenanon, tconcn Crapetedntly,hdrsre n; de aw dncts unchnin rorea eh pvam ="tma 3 exeen aSIC l rm>Fisl orpeeir apps., euablebllyr;; anresues rhat c ir edicati,t appr Wetde frggve012 sm>Fi.justify ilwisem>,wtem>raw tys. Tmedia soem>uot; tem>eoiam payisem>uin bottom-3 e all desieo rais ircumtn; de n ifilcom, seto ba n awnem>lmitt,ntimediadeveloe plpursnbp3wotcoy-haSIC rpeicothltubllyrop scied G- Invd vigtex,eocatio, low-al roinc s/align: a trahm>Figneitp-eseenmak jalign: bs by Seeengaplptttp://ww oronneccarinitsp;&ct qutys. owgtaltlen ifilcom orinhe data ultimebehavamring ethmwrcy, we w dncts ctw lesm ewit er coresues fu sl="textcaicati"ng peeir apps..nirong> G- Prov: Fi: 314.0px;schemustifron.c140,0);"> H-A-H<65pan>">Univernivereilshilso/p>