Research
My research activity deals
with adaptive computing, model based design methodologies,
reconfigurable architectures, rapid system prototyping, embedded system
security
and hardware/software codesign.
Current research topics
- Adaptive computing
- Rapid system prototyping on reconfigurable
platforms
- Software radio on reconfigurable platforms
- Embedded systems security
- Model based design methodologies
Previous research topics
- Communication architecture exploration for MPSoC
systems
- Verification of heterogeneous systems from an
UML specification
- Design space exploration for reconfigurable
architectures
- Performance estimation for FPGAs from a C
specification
- Hardware/software partitioning
- Communication synthesis
- Performance estimation for DSP
Current
Ph.D. Students
- Milad El Khodary,
Operating Environment for Self Adaptive Networked Entities (SANES) (2006 - 2012)
- Pascal Cotret, Distributed
security for communications and memories in a multiprocessor
architecture (2009 - 2012)
- Sébastien Guillet,
Modeling and Formal Control of Partial Dynamic Reconfiguration (2009 -
2012)
- Vianney Lapotre,
Management of reconfigurable multi-standards ASIPs-based receiver (2010
- 2013)
- Simon Fau, Crypto-engines for
homomorphic cryptography (2011-2014)
Ph.D.
Graduates
- Jérémie
Crenne, Embedded system security (2008 - 2011)
now Post-Doc at LIRMM,
Montpellier, France
- Michael
Grand,
Cryptoprocessor for Software Defined Radio
(2008 - 2011)
now Assistant
Professor at
University of Science and Technologies of Bordeaux, France
- Linfeng
Ye, Self-adaptative Multi-processor System-on-Chip :
Architecture, Methodology and Decision (2007 - 2010)
now Assistant Professor
at Guangdong University of Technology, Guangzhou, P.R China
- Jorgiano
vidal, Dynamic and partial reconfigurable embedded systems
design with UML (2007 - 2010)
now Associate Professor
at Federal Tecnological Education Center of Rio Grande do Norte, Brazil
- Rasmus
Abildgren, Implementation Effort and Parallelism - Metrics for
Guiding Hardware/Software Partitioning in Embedded System Design (2006
- 2010)
now a Software Engineer
at CSR, Alborg, Denmark
- Gaël
Abgrall, Software
Defined Radio and dynamic reconfiguration (2007 - thesis interruption
in 2010)
now engineer at
DC-DIRISI, Paris, France
- Yassine
Aoudni, Rapid system
prototyping of reconfigurable platforms (2003
- 2010)
now Teaching Assistant
at ENIS, Sfax, Tunisia
- Romain
Vaslin, Hardware Core for
Off-chip Memory Security Management in Embedded Systems (2005 -
2008)
now an engineer at
Thales Communications, Cholet, France
- Issam
Maalej, Communication
architecture exploration for MPSoC systems (2002 - 2006)
now an R&D engineer
at Sydel, Lorient, France
- Samuel
Rouxel, Modeling and
characterization of heterogeneous SoC platform: Application to SDR (2002 - 2006),
now an R&D engineer
at CRESITT Industrie, Orléans, France
- Lilian Bossuet,
Design space
exploration of
reconfigurable architectures (2001 - 2004),
now an Associate
Professor
at University of Saint-Etienne, France
- Sébastien
Bilavarn, Architectural
exploration from a C specification: FPGAs case study (1999 -
2002),
now an Associate
Professor at University of Nice-Sophia Antipolis, France
M.Sc.
Graduates
- Soukaina Benamar, Design of
Multiprocessor Systems on Chip - 2011
- Said Louizi, Firewall for
communication protection within embedded systems - 2011
- Cedric Seguin, Reconfigurable
architecture and security: how to handle remote partial reconfiguration
securely - 2010
now a
PhD student at
Lab-STICC, Lorient, France
- Naoufel Belfathi,
Multithreading on XilKernel for multiprocessor systems - 2010
- Abdessalam Chafik, Communication
protection within embedded systems - 2010
- Jérémie crenne, Remote
partial reconfiguration through network protocol - 2008,
now a PhD student at
Lab-STICC, Lorient, France
- Yaset Oliva, Bitstream server for
reconfigurable system on chip - 2008,
now a PhD student at
IETR, Rennes, France
- Zui Tao,
Asymetric Encryption Algorithm: Implemntation on Nios-based systems -
2007,
now a Master student at
Université de Bretagne Sud, Lorient, France
- Sylvain Ducloyer,
Hardware architecture for haching functions: Application to
MD5/SHA-1/SHA-2 - 2007
now an engineer at
Lab-STICC, Lorient, France
- Arnaud Dumérat, Fault
detection and fault tolerant ECC algorithm - 2006,
- Jérémie
Guillot, Cryptography
and self dynamic reconfiguration on an FPGA platform – 2004,
now
a Post Doc at
Lab-STICC, Lorient, France
- Jean Philippe
Delahaye, Software radio and
dynamic reconfiguration on a DSP/FPGA platform – 2003,
now an R&D engineer at
CELAR, Rennes, France
- Erwan Piriou,
Reconfigurable technologies and
programmable technologies: a comparison – 2003,
now a R&D engineer at CEA,
Orsay, France
- Samuel Rouxel,
Routing cost on FPGA architectures – 2002,
now an R&D engineer
at CRESITT Industrie, Orléans, France
- Issam Maalej,
Interface synthesis for SoC – 2002,
now an R&D engineer
at Sydel, Lorient, France
- Lilian Bossuet,
Modeling of reconfigurable architectures: toward a generic approach
– 2001,
now an Associate
Professor at University of Science and Technologies of Bordeaux
- Said Chaboun,
Implementation of audio coding onto heterogeneous architectures –
1999
|
Funding
Several fundings support
my research activity: CNRS, RNRT, RNTL, DGA, CMCU,
Brittany State, ANR
Below is a summary of most relevant projects:
- FAMOUS
project (2010-2012): This project aims at introducing a
complete methodology that takes the reconfigurability of the hardware
as an essential design concept and proposes the necessary mechanisms to
fully exploit those capabilities at runtime. The project covers
research in system models, compile time and run time methods, and
analysis and verification techniques. These tools will provide
high-quality designs with improved designer productivity, while
guaranteeing consistency with the initial requirements for adaptability
and the final implementation.
- SecReSoC project (2010-2012): The goals of the
SecReSoC project are to increase the security level of reconfigurable
technologies (FPGAs) at the logic, architectural and system levels.
FPGA technology has been selected as it becomes widespread in many
application domains and corresponds to a strong vector to prototype and
to evaluate the security level of cryptographic architectures. A
generic MPSoC architecture enabling the integration into an FPGA of an
application requiring a multilevel security data protection will be
designed, implemented and tested. This multiprocessor architecture will
include an optimized cryptoprocessor for implementation of
cryptographic protocols and encryption modes; some standard processors
for multitasking OS implementation; internal data memories and an
interface to external data memories; input/output units and an internal
communication structure integrating and enabling several security
levels. Tools for secured device reconfiguration will be proposed also.
System security will be evaluated and side-channel attacks
will be considered during system design and evaluation. The final aim
of the project will be the design of a prototype to evaluate the cost
and the efficiency of the proposed techniques.
- MOPCOM projet (2006 -
2009):
The MOPCOM project is focused on model engineering using MDA approaches
to develop SoC/SoPC. The project aims to provide a formalized
design process (design methodology) and the associated tools in order
to target: The design at different levels of abstraction, from system
TLM level down to architecture RTL level. The design of reconfigurable
systems that may be reconfigured at run time. A MDA/MDE prototype tool
will be developed during the project that will perform code generation
(SystemC and VHDL) from a specification description using UML. This
prototype will rely on adapted profiles for real time embedded system
from the system description down to the architecture description.
- AEther Project (2006 - 2009):
European citizens are now living in a world of "pervasive computing",
where virtually every object has a processing power. Undoubtedly,
computing devices are more ubiquitous and interconnected than ever,
fulfilling the most varied tasks with little human intervention. The
size of these "pervasive computing" networks is significantly
increasing, as well as the variety of the computing devices, both at
chip (multicore and reconfigurable architectures) and system level
(distributed processing). As their scope of application broadens,
processing resources require greater flexibility and scalability to
meet the various needs of users. AETHER's
main objectives are to study, evaluate and propose novel computing
architectures responding to the most demanding embedded
applications in the next 10+ years. In particular, the AETHER project
aims to tackle the issues related to the performance and technological
scalability, increased complexity and programmability of future
embedded computing architectures by introducing self-adaptive
technologies in computing resources.
- ICTeR Project (2006 - 2009):
Digital integrated systems have supplanted the paper as the media
through which the information is transmitted, thus a great interest for
cryptology has emerged at all levels of the integrated circuit design
flow. The physical implementation has indeed become the Achille’s
Heel of secured platforms while side channel attacks such as the
Differential Power Analysis have become common and popular. It is now
well known and accepted that side channel attacks are the most
efficient attacks since they require only little knowledge and material
to be successfully applied. In this context, this project aims at
analysing the potential benefits in terms of security of physical
reconfigurable platforms and devices. More precisely, this projects
aims at (a) defining ad-hoc integration techniques of cryptography
primitives on such platforms and (b) and at defining what is the ideal
reconfigurable platform allowing the best possible integration of
cryptography primitives.
- SANES
Project (2004 - 2005):
Embedded
System Security is becoming a major issue to enable the vision of
ubiquitous computing. Numerous challenges need to be addressed to
promote security within future embedded systems. Our research aims at
defining new solutions to leverage embedded systems security by taking
benefit
of reconfigurable architectures and on-chip hardware monitoring. We
also propose new solutions to improve security within future NoC.
- PROSYR
Project (2003 - 2005):
This project defines a methodology for rapid prototyping of real time
embedded systems. Our approach deals with design space exploration from
system level to physical implementation. The design flow is based on
several steps that enable the designer to progressively refine his
implementation. Metrics at the system level define the orientation of
the application (control, data, memory). Clustering techniques promote
the definition of MPSoC systems. Hardware and software estimation
techniques are also evaluated in order to provide performance
evaluation from a system level specification. RTOS are taken into
account to provide a fine scheduling of the application onto the
heterogeneous platform.
- A3S
Project (2003 - 2005): This
project defines a new methodology for software radio systems
validation, entirely based on UML. Right at the modeling step it will
be
possible to perform non-functional coherence verification of software
radio
architecture specifications and application requirements with UML based
models.
The interest of such an approach is to give the designers the
opportunity to
investigate, before beginning any development step, the array of
potential
solutions and enabling selection of some by verification of the
coherency. This
approach enables design cost saving by drastic reduction of time
and
minimization of the number of prototypes.
- EPICURE
Project (2001 - 2003):
This project defines a new design methodology able to bridge the gap
between an abstract specification and an heterogeneous reconfigurable
architecture. The EPICURE contribution is the result of a joint study
on abstraction/refinement methods and a smart reconfigurable
architecture within the formal Esterel design tools suite. The original
points of this work are: i) a generic HW/SW interface model, ii) a
specification methodology that handles the control, and includes
efficient verification and HW/SW synthesis capabilities, iii) a method
for parallelism exploration based on abstract resources/performance
estimation expressed in terms of area/delay tradeoffs, iv) a HW/SW
partitioning approach that refines the specification into explicit HW
configurations and the associated SW control. The EPICURE framework
shows how a cooperation of complementary methodologies and CAD tools
associated with a relevant architecture can significantly improve the
designer productivity, especially in the context of reconfigurable
architectures.
- MACGTT Project
(2000 - 2002):
This project proposes an original design flow that performs a
progressive design space reduction. In order to fully take benefit of
the optimization potential from the applications and the architectures,
our flow realizes a large parallelism exploration at the algorithmic
level before the definition of the target architecture. Then, based on
the highlighted solutions during the exploration step, an
hardware/software partitionning phase enables the refinement of the
solutions to converge to the characterized final architecture.
|