Framework For Efficient Cosimulation And Fast Prototyping on Multi-Components With AAA Methodology: LAR Codec Study Case

Erwan Flécher (1) , Mickaël Raulet (2) , Ghislain Roquier (1) , Marie Babel (1) , Olivier Déforges (2)

1 IETR - Institut d'Electronique et de Télécommunications de Rennes
2 IETR - Institut d'Électronique et des Technologies du numéRique

Real-time signal and image applications have significant time constraints involving the use of several powerful calculation units. Programmable multi-component architectures have proven to be a suitable solution combining flexibility and computation power. This paper presents a methodology for the fast design of signal and image processing applications. In a unified framework, application modeling, cosimulation and fast implementation onto parallel heterogeneous architectures are enabled and help to reduce time-to-market. Moreover, automatic code generation provides a high abstraction level for users. Finally, the worthwhile nature of Matlab/C language cosimulation is illustrated on a still image codec named LAR.

Domains: Embedded Systems

Erwan Flécher, Mickaël Raulet, Ghislain Roquier, Marie Babel, Olivier Déforges. Framework For Efficient Cosimulation And Fast Prototyping on Multi-Components With AAA Methodology: LAR Codec Study Case. 15th European Signal Processing Conference (Eusipco 2007), Sep 2007, Poznań, Poland. pp.ISBN: 978-83-921340-2-2. ⟨hal-00171799⟩