Title :

On the distribution of clique-based neural networks for edge AI
[Invited]

Author(s) :

Larras, Benoit [Auteur correspondant]
Microélectronique Silicium - IEMN [MICROELEC SI - IEMN]
Institut d’Électronique, de Microélectronique et de Nanotechnologie - UMR 8520 [IEMN]
Frappe, Antoine [Auteur]
Microélectronique Silicium - IEMN [MICROELEC SI - IEMN]
Institut d’Électronique, de Microélectronique et de Nanotechnologie - UMR 8520 [IEMN]

Journal title :

IEEE Journal on Emerging and Selected Topics in Circuits and Systems

Pages :

469-477

Publisher :

IEEE

Publication date :

2020-12

ISSN :

2156-3357

English keyword(s) :

Neural networks
Wireless sensor networks
Intelligent sensors
Energy consumption
Monitoring
Feature extraction
Neural networks circuit
clique-based neural networks
analog
mixed-signal circuit
distributed architecture

HAL domain(s) :

Sciences de l'ingénieur [physics]/Micro et nanotechnologies/Microélectronique
Informatique [cs]/Systèmes embarqués
Informatique [cs]/Intelligence artificielle [cs.AI]
Informatique [cs]/Réseau de neurones [cs.NE]

English abstract : [en]

Distributed smart sensors are more and more used in applications such as biomedical or domestic monitoring. However, each sensor broadcasts data wirelessly to the others or to an aggregator, which leads to energy-hungry ...
Show more >Distributed smart sensors are more and more used in applications such as biomedical or domestic monitoring. However, each sensor broadcasts data wirelessly to the others or to an aggregator, which leads to energy-hungry sensor nodes not ensuring data privacy. To tackle both challenges, this work proposes to distribute the feature extraction and a part of a clique-based neural network (CBNN) in each sensor node. This scheme allows standardizing data at the sensor level, ensuring privacy if the data is intercepted. Besides, a lower number of bits is transmitted, thus limiting the communication overhead. The inherent redundancy of clique-based networks makes them resilient to out-of-range connections, allowing an additional power reduction in the sensor nodes. Compared with a localized CBNN in the aggregator, the distributed structure reduces the inference latency by 28%, the sensor energy consumption by 25% and increases the protocol robustness. The circuit implementation is possible with the use of single-cluster iterative clique-based circuits, and demonstrated for a posture recognition application. To this end, a hardware circuit has been fabricated and performs a classification using 115fJ per synaptic event per neuron in 83ns.Show less >

Language :

Anglais

Popular science :

Non

ANR Project :

Event Driven Artificial Intelligence Hardware for Biomedical Sensors

Collections :

• Institut d'Électronique, de Microélectronique et de Nanotechnologie (IEMN) - UMR 8520

Source :

Harvested from HAL