Section: Partnerships and Cooperations

National Initiatives

The Cairn team mainly collaborates with the following laboratories: CEA List, CEA Leti, LEAT Nice, Lab-Sticc (Lorient, Brest), LIRMM (Montpellier, Perpignan), LIP6 Paris, IETR Rennes, DTIM-ONERA Toulouse, LAAS Toulouse, IRIT Toulouse, Inria Socrate.

The team participates in the activities of the following research organization of CNRS (GdR for in French "Groupe de Recherche"):

• GdR SOC-SIP (System On Chip & System In Package), working groups on reconfigurable architectures, embedded software for SoC, low power issues. E. Casseau is in charge of the architecture topic of the reconfigurable platform working group.

• GdR ISIS (Information Signal ImageS), working group on Algorithms Architectures Adequation.

• GdR ASR (Architectures Systèmes et Réseaux)

• GdR IM (Informatique Mathématiques), C2 working group on Codes and Cryptography and ARITH working group on Computer Arithmetic

ANR Blanc - PAVOIS (2012–2016)

Participants : Arnaud Tisserand, Emmanuel Casseau, Philippe Quémerais, Jérémie Métairie, Nicolas Veyrat-Charvillon, Karim Bigou, Pierre Guilloux.

PAVOIS (in French: Protections Arithmétiques Vis à vis des attaques physiques pour la cryptOgraphIe basée sur les courbeS elliptiques) is a project on Arithmetic Protections Against Physical Attacks for Elliptic Curve based Cryptography. It involves IRISA-Cairn (Lannion) and LIRMM (Perpignan and Montpellier). This project will provide novel implementations of curve based cryptographic algorithms on custom hardware platforms. A specific focus will be placed on trade-offs between efficiency and robustness against physical attacks. One of our goal is to theoretically study and practically measure the impact of various protection schemes on the performance (speed, silicon cost and power consumption). Theoretical aspects will include an investigation of how special number representations can be used to speed-up cryptographic algorithms, and protect cryptographic devices from physical attacks. On the practical side, we will design innovative cryptographic hardware architectures of a specific processor based on the theoretical advancements described above to implement curve based protocols. We will target efficient and secure implementations for both FPGA an ASIC circuits. For more details see http://pavois.irisa.fr .

ANR INFRA 2011 - FAON (2012-2015)

Participants : Raphaël Bardoux, Arnaud Carer.

The FAON (Frequency based Access Optical Networks) project objectives are to demonstrate the technology and feasibility of a new type of Passive Optical Network (PON) for broadband access which uses a Frequency based shared access technique known as Frequency Division Multiplexing (FDM). These goals completely fall into the line of the expected capacity increase in PON which is today forecasted to go from 100 Mbps per user to 1 Gbps. Faon involves Orange Labs, CEA-LETI, University of South Brittany (Lab-STICC laboratory) and Univ. Rennes I (Foton laboratory and Cairn team). Cairn developed a high-rate architecture at the receiver side. Specific receiver algorithms (synchronization and equalization) and FPGA implementation are the key issues that were addressed. This project ended in 2015.

ANR Ingénérie Numérique et Sécurité - ARDyT (2011-2016)

Participants : Arnaud Tisserand, Philippe Quémerais.

ARDyT (in French: Architecture Reconfigurable Dynamiquement Tolérante aux fautes) is a project on a Reliable and Reconfigurable Dynamic Architecture. It involves IRISA-Cairn (Lannion), Lab-STICC (Lorient), LIEN (Nancy) and ATMEL. The purpose of the ARDyT project is to provide a complete environment for the design of a fault tolerant and self-adaptable platform. Then, a platform architecture, its programming environment and management methodologies for diagnosis, testability and reliability have to be defined and implemented. The considered techniques are exempt from the use of hardened components for terrestrial and aeronautics applications for the design of low-cost solutions. The ARDyT platform will provide a European alternative to import ITAR constraints for fault-tolerant reconfigurable architectures. For more details see http://ardyt.irisa.fr .

ANR Ingénérie Numérique et Sécurité - COMPA (2011-2015)

Participants : Emmanuel Casseau, Steven Derrien, Yaset Oliva Venegas.

COMPA (model oriented design of embedded and adaptive multiprocessor) is a project which involves Cairn , IETR (Rennes) and Lab-STICC (Lorient). The aim of the project was to design adaptive multiprocessor embedded systems for executing dataflow programs. The use case is the Reconfigurable Video Coding (RVC) standard. More specifically, we focus on the portable and platform-independent RVC-CAL language to describe the applications. We use transformations to refine, increase parallelism and translate the application model into software and hardware components. Specific scheduling and actor's mapping are also investigated for runtime execution. For more details see http://www.compa-project.org . This project ended in 2015.

ANR Ingénérie Numérique et Sécurité - DEFIS (2011-2015)

Participants : Olivier Sentieys, Nicolas Simon.

DEFIS (Design of fixed-point embedded systems) is a project which involves Cairn , LIP6 (University of Paris 6), LIRMM (University of Perpignan), CEA LIST, Thales, Inpixal. The main objectives of the project were to propose new approaches to improve the efficiency of the floating-point to fixed-point conversion process and to provide a complete design flow for fixed-point refinement of complex applications. This infrastructure reduces the time-to-market by automating the fixed-point conversion and by mastering the trade-off between application quality and implementation cost. Moreover, this flow guarantees and validates the numerical behavior of the resulting implementation. The proposed infrastructure was validated on two real applications provided by the industrial partners. For more details see http://defis.lip6.fr . This project ended in 2015.

Labex CominLabs - BoWI (2012-2016)

Participants : Olivier Sentieys, Arnaud Carer.

The BoWi project (Body Wold Interactions) aims at designing an accurate gesture and body movement estimation using very-small and low-power wearable sensor nodes. It initially stems from a proposal of the CominLabs think thank focused on the society challenge called Digital Environment for the Citizen. It is also related to the social challenge ICT for Personalized Medicine and to the research track Energy Efficiency in ICT. The main objective of the project is to propose pioneer interfaces for an emerging interacting world based on smart environments (house, media, information and entertainment systems...). Basically the project relies on Wireless Body Areas Sensor Networks; the aim is the accurate Gesture and Body Movement estimation with extremely severe constraints in terms of footprint and power consumption according to on-body energy harvesting perspectives. The BoWI geolocation approach will combine radio communication distance measurement and inertial sensors and it will also strongly benefit from cooperative techniques based on multiple observations and distributed computation. Different types of applications, as health care, activity monitoring and environment control, will be considered and evaluated along with a human-machine interface expertise.

The scientific challenge is global and deals with the solution to be interactively invented by all partners: a short-range geolocation method based on distributed and coopering devices processing multisource data issued from radio-communication distance estimation and integrated inertial sensors. It includes several specific contributions:

• Dynamic and cooperative communication coding and protocol for inter-nodes communications. This includes cooperative communications and protocols such as cooperative MIMO, relaying, error coding, network coding and MAC and wake-up radio protocols.

• Node hardware/software architecture design and self-adaptive distributed processing for geolocation with aggressive low-power run-time optimisation.

• Channel models and antennas for short-range communications. This study will be performed for various radio standards from upcoming BAN 802.15.6, 802.15.4a technologies to future UWB solutions.

• Channel models and antennas for WBASN at millimeter waves. This is a promising perspective for antenna miniaturization, however no front-ends are yet available.

• In depth and specific analysis of human-machine interactions to set system constrains and define user requirement according to various application perspectives.

In practice the BoWi partners aim to deliver the design of basic components, a prototype based on available radio front-ends and energy harvesting devices as well as a system simulator including mm-wave models. Results will also concern the specification of future radio-front ends. The BoWI involves Cairn , IRISA Granit (Lannion), IETR (Rennes), and Lab-STICC (Brest, Lorient, Vannes). For more details see http://www.bowi.cominlabs.ueb.eu/fr .

Labex CominLabs - 3DCORE (2014-2018)

Participants : Olivier Sentieys, Daniel Chillet, Cédric Killian, Jiating Luo, Van Dung Pham.

3DCORE (3D Many-Core Architectures based on Optical Network on Chip) is a project which involves Cairn , FOTON (Rennes, Lannion) and Institut des Nanotechnologies de Lyon. 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. The 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) contributes 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 photonics layer suitable for a flexible and energy efficient high-speed optical network on chip (ONoC).

Labex CominLabs - RELIASIC (2014-2018)

Participants : Emmanuel Casseau, Arnaud Tisserand.

RELIASIC (Reliable Asic) is a project which involves Cairn , Lab-STICC (University of Bretagne Sud) and IETR (Institut d'Electronique et de Télécommunications de Rennes). One of the most critical challenges of the next design technologies will be 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. New techniques will be required to increase the reliability of operators and components. The aim of the project is to address this problem with a bottom-up approach, starting from an existing application as a use case (a GPS receiver) and adding some redundant mechanisms to allow the GPS receiver to be tolerant to transient errors due to low voltage supply.

Labex CominLabs & Lebesgue - H-A-H (2014-2017)

Participants : Arnaud Tisserand, Nicolas Veyrat-Charvillon, Karim Bigou, Gabriel Gallin.

H-A-H for Hardware and Arithmetic for Hyperelliptic Curves Cryptography is a project on advanced arithmetic representation and algorithms for hyper-elliptic curve cryptography. It involves IRISA-Cairn (Lannion) and IRMAR (Rennes).

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). This project will provide novel implementations of HECC based cryptographic algorithms on custom hardware platforms. For more details see http://h-a-h.inria.fr/ .