Descripción del proyecto
Aceleradores neuromórficos seguros y energéticamente eficientes basados en la fotónica de silicio aumentada
La computación neuromórfica es un método inspirado en el funcionamiento del cerebro para procesar grandes cantidades de datos con una elevada eficiencia energética, una baja latencia y un gran ancho de banda. Los sistemas neuromórficos son muy atractivos para las aplicaciones de computación perimetral gracias a su elevado rendimiento y ligereza. Sin embargo, en este tipo de sistemas, sobre todo en aplicaciones críticas para la seguridad, es obligatorio la presencia de capas de seguridad sólidas y energéticamente eficientes, ya que es muy fácil que se vean comprometidas en el borde. El objetivo del proyecto NEUROPULS, financiado con fondos europeos, es crear sistemas de computación perimetral seguros y de bajo consumo mediante el desarrollo de nuevas arquitecturas informáticas fotónicas, así como de capas de seguridad basadas en funciones fotónicas físicas no clonables en plataformas fotónicas de silicio aumentadas que sean compatibles con la tecnología de semiconductores complementarios de óxido metálico. El equipo de NEUROPULS desarrollará nuevas plataformas tecnológicas, de «hardware» y de simulación para crear aceleradores neuromórficos de nueva generación con interfaces compatibles con RISC-V para, de este modo, facilitar la compatibilidad de los sistemas.
Objetivo
The growing need to transfer massive amounts of data among multitudes of interconnected devices for e.g. self-driving vehicles, IoT or industry 4.0 has led to a quest towards low-power and secure approaches to locally processing data. Neuromorphic computing, a brain-inspired approach, addresses this need by radically changing the processing of information. Although neuromorphic electrical computing systems offer advantages in terms of CMOS implementations and scalability, they inherit limitations of conventional electronics such as low energy-efficiency, high latency and low bandwidth density. Besides, such systems often require robust security layers for e.g. safety-critical applications. Security layers based on memory-stored secret keys are prone to several types of memory-accessing attacks. Therefore, silicon hardware approaches for security primitives such as physical unclonable functions (PUFs) are currently investigated because of their absence of long-term digital memory storage. Although electronic PUFs have received major attention thanks to their native CMOS implementation, for secure authentication they are prone to machine learning and side-channel attacks due to their CMOS technology.
The NEUROPULS project aims to build next-generation low-power and secure edge-computing systems by developing novel photonic computing architectures and security layers based on photonic PUFs in augmented silicon photonics CMOS-compatible platforms. The integration of emerging non-volatile phase change materials for synapses/neurons and III-V materials for on-chip spiking sources, for the first time, will allow to build novel neuromorphic accelerators featuring RISC-V compliant interfaces for smooth adoption and programmability. Optimal performance will be achieved thanks to a novel full-system simulation platform for design space exploration. Three relevant use-cases will be considered for benchmarking to demonstrate 2 orders of magnitude energy efficiency improvement.
Ámbito científico
- natural sciencescomputer and information sciencescomputer securitycryptography
- natural sciencescomputer and information sciencescomputational sciencemultiphysics
- natural sciencescomputer and information sciencesartificial intelligencemachine learning
- engineering and technologynanotechnologynanophotonics
- natural sciencescomputer and information sciencessoftwaresoftware applicationssimulation software
Palabras clave
Programa(s)
Convocatoria de propuestas
HORIZON-CL4-2021-DIGITAL-EMERGING-01
Consulte otros proyectos de esta convocatoriaRégimen de financiación
HORIZON-RIA - HORIZON Research and Innovation ActionsCoordinador
75794 Paris
Francia