DSP on programmable component

Since the early 2000s, reconfigurable hardware components such as FPGAs (Field Programmable Gate Arrays) have been real alternatives to ASICs (Application Specific Integrated Circuits). In the field of signal processing, exploiting the hardware parallelism of FPGAs makes it possible to achieve higher performance than conventional implementations targeting DSPs (Digital Signal Processors). To increase the suitability of FPGAs for signal processing, circuit manufacturers have upgraded their products both architecturally (appearance of hardware arithmetic blocks dedicated to signal processing) and software-wise (development of high-level design tools linked to algorithm development tools such as Matlab). Today's FPGAs are therefore key components for implementing signal processing applications.


This course aims to provide the knowledge and methods needed to efficiently implement signal processing applications using FPGA-type circuits.

- Knowledge of signal processing and digital systems design (see Rémi Megret course and Xavier Miet course)
- Basics of VHDL programming (see Patrice Nouel - Sylvie Renaud course)
- Basics of Matlab programming

- Chapter I: Introduction
o Overview of digital system design
o Overview of signal processing requirements

- Chapter II: FPGA technology, architecture and use

- Chapter III: The benefits of FPGAs for signal processing

- Chapter IV: Overview of signal processing application design tools for FPGAs

- Chapter V: Case studies

- Chapter VI: Evolving the field: reconfigurable architectures

- Application - Practical work