Sub-0.3V CMOS neuromorphic technology and its potential application

Archive ouverte : Communication dans un congrès

Danneville, Francois | Carpentier, Kevin | Sourikopoulos, Ilias | Paindavoine, Michel | Loyez, Christophe

Edité par HAL CCSD ; IEEE

oral special session. International audience. The aim of this paper is to present a sub-0.3 V neuromorphic technology developed for spiking neural network design and its potential application. The main properties of the developed ultra low power (ULP) artificial neuron are first recalled. A description of ULP synapses follows that includes the plasticity scheme. The neuromorphic toolbox is then used to design a basic circuit allowing oriented edges classification. The circuit is made of 40 neurons and 108 plastic synapses, its consumed silicon core area is 0.025 mm 2 and the overall power consumption of 5 nW. Finally, the deployment of the technology within an industrial context to fabricate highly energy efficient Spike-based visual sensor is discussed.

Consulter en ligne

Suggestions

Du même auteur

Subthreshold neuromorphic devices for spiking neural networks applied to embedded A.I | Loyez, Christophe

Subthreshold neuromorphic devices for spiking neural networks applied to em...

Archive ouverte: Communication dans un congrès

Loyez, Christophe | 2021-06-13

International audience. Energy autonomy is one of the major challenges of embedded Artificial Intelligence. Among the candidate technologies likely to take up such a challenge, spiking neural networks are the most p...

Bioinspired frugal technology for biomimetic optical sensors | Danneville, Francois

Bioinspired frugal technology for biomimetic optical sensors

Archive ouverte: Communication dans un congrès

Danneville, Francois | 2022-02-25

oral. National audience

Coarse/fine delay element design in 28 nm FD-SOI | Sourikopoulos, Ilias

Coarse/fine delay element design in 28 nm FD-SOI

Archive ouverte: Type de document indéfini

Sourikopoulos, Ilias | 2020-04-26

International audience. This chapter starts by highlighting the requirements and existing techniques in producing digital delay, summarizing the state of the art. Subsequently, a topology assessment is presented bas...

Du même sujet

Event-Driven Continuous-Time Feature Extraction for Ultra Low-Power Audio Keyword Spotting | Mourrane, S.

Event-Driven Continuous-Time Feature Extraction for Ultra Low-Power Audio K...

Archive ouverte: Communication dans un congrès

Mourrane, S. | 2021-06-06

International audience. In the context of autonomous keyword spotting and sound detection, this paper proposes a low power feature extraction unit generating spectrograms that represent a unique signature allowing t...

Voltage-dependent synaptic plasticity: Unsupervised probabilistic Hebbian plasticity rule based on neurons membrane potential | Garg, Nikhil

Voltage-dependent synaptic plasticity: Unsupervised probabilistic Hebbian p...

Archive ouverte: Article de revue

Garg, Nikhil | 2022-10-21

International audience. This study proposes voltage-dependent-synaptic plasticity (VDSP), a novel brain-inspired unsupervised local learning rule for the online implementation of Hebb’s plasticity mechanism on neuro...

A Sub-100 µW 0.1-to-27 Mb/s Pulse-based Digital Transmitter for the Human Intranet in 28 nm FD-SOI CMOS | Tochou, Guillaume

A Sub-100 µW 0.1-to-27 Mb/s Pulse-based Digital Transmitter for the Human I...

Archive ouverte: Article de revue

Tochou, Guillaume | 2022-05

International audience. Human body communications require energyefficient transceivers to connect diverse devices on the human body for wellness and medical applications. This paper presents a fully digital pulse-ba...

A Fully-Digital 0.1-to-27 Mb/s ULV 450 MHz Transmitter with sub-100 µW Power Consumption for Body-Coupled Communication in 28 nm FD-SOI CMOS | Tochou, Guillaume

A Fully-Digital 0.1-to-27 Mb/s ULV 450 MHz Transmitter with sub-100 µW Powe...

Archive ouverte: Communication dans un congrès

Tochou, Guillaume | 2021-06-07

International audience. A 0.5 V fully-digital 450 MHz transmitter for surface wave capacitive body-coupled communications is realized in 28 nm FD-SOI CMOS and consumes 17 to 76 µW for data rates from 0.1 to 27 Mb/s ...

Securing Deep Spiking Neural Networks against Adversarial Attacks through Inherent Structural Parameters | El-Allami, R.

Securing Deep Spiking Neural Networks against Adversarial Attacks through I...

Archive ouverte: Communication dans un congrès

El-Allami, R. | 2021-02-01

International audience. Deep Learning (DL) algorithms have gained popularity owing to their practical problem-solving capacity. However, they suffer from a serious integrity threat, i.e., their vulnerability to adve...

Chargement des enrichissements...