Maître de Conférences
I am an associate professor at the National Institute of Applied Sciences (INSA) in Rennes. I hold a joint research appointment at IETR in Rennes and at Institut Pascal in Clermont Ferrand, two CNRS research units. I work on embedded systems design, image processing and parallel systems programming, with particular interests in model-based design, dataflow models of computation, and models of architecture.
Overall, I'm thrilled to discuss the challenges of efficient and secure processing, in particular for signal processing and machine learning powered applications such as in telecommunications, video compression and computer vision.
Design of Heterogeneous Distributed Embedded Systems, Parallel Digital Signal Processing, Models of Computation, Models of Architecture, Low Energy Processing, Image and Telecommunication Processing, FPGAs, MPSoCs, many-cores, Approximate Computing, Cybersecurity, Embedded Deep Learning.
Associate at Institut Pascal, DREAM team.
Main Projects I am involved in
Description: The CERBERO EU project aims at developing a design environment for Cyber-Physical Systems based on two pillars:
1. a cross-layer model-based approach to describe, optimize, and analyze the system and all its different views concurrently, and
2. an advanced adaptivity support based on a multilayer autonomous engine.
The targeted three applications of CERBERO are a self-healing system for planetary exploration, unmanned vehicles for ocean monitoring and a smart travel network for electric vehicles.
Description: The COMPACT-SL-MODELS project, financed by US National Science Foundation, develops new techniques to help advanced computing systems for signal processing better adapt to the environments in which they operate.
Description: The aim of the ANR HPeC project is to develop self-adaptive, energy-efficient high performance embedded computing in the context of Unmanned Aerial Vehicles.
Description: The aim or the FUI GreenVIDEO project is to study and develop mechanisms for reducing energy consumption of image/video creation, transformation, broadcast and rendering.
Description: The aim of the ANR COMPA project is to propose a methodology and a design framework for running applications specified with dataflow process networks on a heterogeneous distributed embedded system.
- Industrial Collaboration with Texas Instruments France on multicore DSP programming.
Main Developped Tool
PREESM is an open source dataflow-based system synthesizer. It is primarily employed to simulate signal processing applications and to generate code for multi-core processors.
Main Used Tools
CAPH Compiler is an open source compiler for programming FPGAs with a high level dataflow representation.
Orcc is an open source tool for multi-purpose dynamic dataflow programming of parallel systems.
GP Studio is an open source tool for the management of smart cameras.
Institut National des Sciences Appliquées Rennes (INSA Rennes)
Department of Electronics and Computer Engineering (EII)
Microcontroller Programming, Embedded Linux, Hardware Design, VHDL, Operating Systems, Distributed Programming, Analog Electronics, Logic, Multidisciplinary Projects.
As an example of a teaching multidisciplinary project, ~chos (pronounce "echos") is a combined Art and Science project conducted with the artists Haythem Zakaria and Yro Yto experimented in 2014 with a class of 50 Master 1 students.