A Noise-Resilient Neuromorphic Digit Classifier Based on NOR Flash Memories with Pulse–Width Modulation Scheme
In this work, we investigate the implementation of a neuromorphic digit classifier based on NOR Flash memory arrays as artificial synaptic arrays and exploiting a pulse-width modulation (PWM) scheme. Its performance is compared in presence of various noise sources against what achieved when a classi...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-11-01
|
Series: | Electronics |
Subjects: | |
Online Access: | https://www.mdpi.com/2079-9292/10/22/2784 |
_version_ | 1797510509275119616 |
---|---|
author | Gerardo Malavena Alessandro Sottocornola Spinelli Christian Monzio Compagnoni |
author_facet | Gerardo Malavena Alessandro Sottocornola Spinelli Christian Monzio Compagnoni |
author_sort | Gerardo Malavena |
collection | DOAJ |
description | In this work, we investigate the implementation of a neuromorphic digit classifier based on NOR Flash memory arrays as artificial synaptic arrays and exploiting a pulse-width modulation (PWM) scheme. Its performance is compared in presence of various noise sources against what achieved when a classical pulse-amplitude modulation (PAM) scheme is employed. First, by modeling the cell threshold voltage (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>V</mi><mi>T</mi></msub></semantics></math></inline-formula>) placement affected by program noise during a program-and-verify scheme based on incremental step pulse programming (ISPP), we show that the classifier truthfulness degradation due to the limited program accuracy achieved in the PWM case is considerably lower than that obtained with the PAM approach. Then, a similar analysis is carried out to investigate the classifier behavior after program in presence of cell <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>V</mi><mi>T</mi></msub></semantics></math></inline-formula> instabilities due to random telegraph noise (RTN) and to temperature variations, leading again to results in favor of the PWM approach. In light of these results, the present work suggests a viable solution to overcome some of the more serious reliability issues of NOR Flash-based artificial neural networks, paving the way to the implementation of highly-reliable, noise-resilient neuromorphic systems. |
first_indexed | 2024-03-10T05:33:22Z |
format | Article |
id | doaj.art-67e11673a39745f2bf3d6269ba7356f6 |
institution | Directory Open Access Journal |
issn | 2079-9292 |
language | English |
last_indexed | 2024-03-10T05:33:22Z |
publishDate | 2021-11-01 |
publisher | MDPI AG |
record_format | Article |
series | Electronics |
spelling | doaj.art-67e11673a39745f2bf3d6269ba7356f62023-11-22T23:07:04ZengMDPI AGElectronics2079-92922021-11-011022278410.3390/electronics10222784A Noise-Resilient Neuromorphic Digit Classifier Based on NOR Flash Memories with Pulse–Width Modulation SchemeGerardo Malavena0Alessandro Sottocornola Spinelli1Christian Monzio Compagnoni2Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano, 20133 Milan, ItalyDipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano, 20133 Milan, ItalyDipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano, 20133 Milan, ItalyIn this work, we investigate the implementation of a neuromorphic digit classifier based on NOR Flash memory arrays as artificial synaptic arrays and exploiting a pulse-width modulation (PWM) scheme. Its performance is compared in presence of various noise sources against what achieved when a classical pulse-amplitude modulation (PAM) scheme is employed. First, by modeling the cell threshold voltage (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>V</mi><mi>T</mi></msub></semantics></math></inline-formula>) placement affected by program noise during a program-and-verify scheme based on incremental step pulse programming (ISPP), we show that the classifier truthfulness degradation due to the limited program accuracy achieved in the PWM case is considerably lower than that obtained with the PAM approach. Then, a similar analysis is carried out to investigate the classifier behavior after program in presence of cell <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>V</mi><mi>T</mi></msub></semantics></math></inline-formula> instabilities due to random telegraph noise (RTN) and to temperature variations, leading again to results in favor of the PWM approach. In light of these results, the present work suggests a viable solution to overcome some of the more serious reliability issues of NOR Flash-based artificial neural networks, paving the way to the implementation of highly-reliable, noise-resilient neuromorphic systems.https://www.mdpi.com/2079-9292/10/22/2784artificial neural networksneuromorphic computingNOR Flash memory arraysprogram noiserandom telegraph noisepulse-width modulation |
spellingShingle | Gerardo Malavena Alessandro Sottocornola Spinelli Christian Monzio Compagnoni A Noise-Resilient Neuromorphic Digit Classifier Based on NOR Flash Memories with Pulse–Width Modulation Scheme Electronics artificial neural networks neuromorphic computing NOR Flash memory arrays program noise random telegraph noise pulse-width modulation |
title | A Noise-Resilient Neuromorphic Digit Classifier Based on NOR Flash Memories with Pulse–Width Modulation Scheme |
title_full | A Noise-Resilient Neuromorphic Digit Classifier Based on NOR Flash Memories with Pulse–Width Modulation Scheme |
title_fullStr | A Noise-Resilient Neuromorphic Digit Classifier Based on NOR Flash Memories with Pulse–Width Modulation Scheme |
title_full_unstemmed | A Noise-Resilient Neuromorphic Digit Classifier Based on NOR Flash Memories with Pulse–Width Modulation Scheme |
title_short | A Noise-Resilient Neuromorphic Digit Classifier Based on NOR Flash Memories with Pulse–Width Modulation Scheme |
title_sort | noise resilient neuromorphic digit classifier based on nor flash memories with pulse width modulation scheme |
topic | artificial neural networks neuromorphic computing NOR Flash memory arrays program noise random telegraph noise pulse-width modulation |
url | https://www.mdpi.com/2079-9292/10/22/2784 |
work_keys_str_mv | AT gerardomalavena anoiseresilientneuromorphicdigitclassifierbasedonnorflashmemorieswithpulsewidthmodulationscheme AT alessandrosottocornolaspinelli anoiseresilientneuromorphicdigitclassifierbasedonnorflashmemorieswithpulsewidthmodulationscheme AT christianmonziocompagnoni anoiseresilientneuromorphicdigitclassifierbasedonnorflashmemorieswithpulsewidthmodulationscheme AT gerardomalavena noiseresilientneuromorphicdigitclassifierbasedonnorflashmemorieswithpulsewidthmodulationscheme AT alessandrosottocornolaspinelli noiseresilientneuromorphicdigitclassifierbasedonnorflashmemorieswithpulsewidthmodulationscheme AT christianmonziocompagnoni noiseresilientneuromorphicdigitclassifierbasedonnorflashmemorieswithpulsewidthmodulationscheme |