NEUROPULS Project: Low-power, secure edge computing systems

NEUROmorphic energy-efficient secure accelerators based on Phase change materials aUgmented siLicon photonicS

Neuromorphic computing is a brain-inspired approach for processing large amounts of data with high energy efficiency, low latency, and high bandwidth.

Neuromorphic systems are particularly attractive for edge computing applications thanks to their high performance and lightweight features. However, strong and energy-efficient security layers are mandatory for such systems, especially for safety-critical applications, as they are very easily compromised at the edge.

Low-power, secure edge computing systems

The EU-funded NEUROPULS project aims to build low-power, secure edge computing systems by developing novel photonic computing architectures and security layers based on photonic physical unclonable functions in augmented silicon photonics platforms compatible with CMOS technology.

By developing novel technological, hardware, and simulation platforms, NEUROPULS aims to create next-generation neuromorphic accelerators featuring RISC-V-compliant interfaces for easier system compatibility.

Project information

PROJECT NAME: NEUROPULS:

START/END DATE: 1/1/2023 – 31/12/2026

KEY THEMES: neuromorphic computing, integrated photonics, security, simulations, RISC-V, CMOS-compatible platforms

PARTNERS: France: Centre national de la recherche scientifique (coordinator), Ecole Centrale de Lyon, Université de Bourgogne, Commissariat à l’Energie atomique et aux energies alternatives; Belgium: Ghent University (UGENT), Hewlett Packard Enterprise; Czechia: Argotech; Germany ; Ludwig-Maximilians-Universität Muenchen, Technische Universität Berlin; Greece: National and Kapodistrian University of Athens; Italy: Politecnico di Torino, Universita degli Studi di Verona; Portugal: Instituto de Engenharia de Sistemas e Computadores, Investigacao e Desenvolvimento Em Lisboa; Spain: Barcelona Supercomputing Center, Albora Technologies.

BUDGET: € 8.3 million