An Adaptive Read Control Voltage Scheme for Reliability Enhancement of Flash-Based In-Memory Computing Architecture for Neural Network
The storage reliability is critical for flash memory based computing in-memory (CIM) architecture for Convolutional Neural Network (CNN). In this paper, we constructed a CIM scheme based on the Nor Flash array (NFA). We conducted simulations to investigate the impact of threshold voltage distributio...
Gespeichert in:
| Veröffentlicht in: | IEEE transactions on device and materials reliability 2024-09, Vol.24 (3), p.422-427 |
|---|---|
| Hauptverfasser: | , , , , |
| Format: | Magazinearticle |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 427 |
|---|---|
| container_issue | 3 |
| container_start_page | 422 |
| container_title | IEEE transactions on device and materials reliability |
| container_volume | 24 |
| creator | Zhang, Xinrui Huang, Jian Liu, Xianping Zhong, Baiqing Yu, Zhiyi |
| description | The storage reliability is critical for flash memory based computing in-memory (CIM) architecture for Convolutional Neural Network (CNN). In this paper, we constructed a CIM scheme based on the Nor Flash array (NFA). We conducted simulations to investigate the impact of threshold voltage distribution and drift of Flash memory cells on the recognition accuracy for various CNN architectures based on the CIM schemes. Based on the reliability study, we proposed a novel compensation scheme to effectively mitigate the impact of threshold voltage drift and evaluated the effectiveness of the proposed scheme by recognition accuracy evaluation. |
| doi_str_mv | 10.1109/TDMR.2024.3429662 |
| format | Magazinearticle |
| fullrecord | <record><control><sourceid>proquest_RIE</sourceid><recordid>TN_cdi_ieee_primary_10601194</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>10601194</ieee_id><sourcerecordid>3101348813</sourcerecordid><originalsourceid>FETCH-LOGICAL-c176t-3dc6e6a67df03920b71efd6c3b9759ae46f4459d515c7c5bc8483a0cd90804de3</originalsourceid><addsrcrecordid>eNpNkN1OgzAUx4nRxDl9ABMvmnjNbGkpcIlz0yVTkzm9JaU9DCZQLEWzF_C5hbALr85J_h8n5-c41wTPCMHR3fbheTPzsMdmlHkR596JMyG-H7qeH7DTYafYZTQMz52Ltt1jTKLA5xPnN65RrERji29AGxAKzXVtjS7Rhy6t2AF6kzlUgDJter0sRFqUhT2gRZ2LWvZKbZHO0LIUbe7eixYUWtXuM1TaHPququlsUe9QbGReWJC2M2PXC3RGlP2wP9p8XjpnmShbuDrOqfO-XGznT-769XE1j9euJAG3LlWSAxc8UBmmkYfTgECmuKRp_00kgPGMMT9SPvFlIP1UhiykAksV4RAzBXTq3I69jdFfHbQ22evO1P3JhBJMKAtDQnsXGV3S6LY1kCWNKSphDgnByYA7GXAnA-7kiLvP3IyZAgD--TkmJGL0D3J0fSc</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>magazinearticle</recordtype><pqid>3101348813</pqid></control><display><type>magazinearticle</type><title>An Adaptive Read Control Voltage Scheme for Reliability Enhancement of Flash-Based In-Memory Computing Architecture for Neural Network</title><source>IEEE Xplore</source><creator>Zhang, Xinrui ; Huang, Jian ; Liu, Xianping ; Zhong, Baiqing ; Yu, Zhiyi</creator><creatorcontrib>Zhang, Xinrui ; Huang, Jian ; Liu, Xianping ; Zhong, Baiqing ; Yu, Zhiyi</creatorcontrib><description>The storage reliability is critical for flash memory based computing in-memory (CIM) architecture for Convolutional Neural Network (CNN). In this paper, we constructed a CIM scheme based on the Nor Flash array (NFA). We conducted simulations to investigate the impact of threshold voltage distribution and drift of Flash memory cells on the recognition accuracy for various CNN architectures based on the CIM schemes. Based on the reliability study, we proposed a novel compensation scheme to effectively mitigate the impact of threshold voltage drift and evaluated the effectiveness of the proposed scheme by recognition accuracy evaluation.</description><identifier>ISSN: 1530-4388</identifier><identifier>EISSN: 1558-2574</identifier><identifier>DOI: 10.1109/TDMR.2024.3429662</identifier><identifier>CODEN: ITDMA2</identifier><language>eng</language><publisher>New York: IEEE</publisher><subject>Accuracy ; Adaptive control ; Artificial neural networks ; Computation ; Computing-in-memory ; convolutional neural network ; Convolutional neural networks ; Drift ; Flash memories ; Flash memory (computers) ; flash memory reliability ; In-memory computing ; multi-level NOR-flash ; Network reliability ; Recognition ; Reliability ; Reliability analysis ; Threshold voltage ; Transistors</subject><ispartof>IEEE transactions on device and materials reliability, 2024-09, Vol.24 (3), p.422-427</ispartof><rights>Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2024</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c176t-3dc6e6a67df03920b71efd6c3b9759ae46f4459d515c7c5bc8483a0cd90804de3</cites><orcidid>0009-0008-3110-5223 ; 0009-0003-1106-4616 ; 0009-0002-3301-6861 ; 0000-0002-8802-0457</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/10601194$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>776,780,792,27902,54733</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/10601194$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Zhang, Xinrui</creatorcontrib><creatorcontrib>Huang, Jian</creatorcontrib><creatorcontrib>Liu, Xianping</creatorcontrib><creatorcontrib>Zhong, Baiqing</creatorcontrib><creatorcontrib>Yu, Zhiyi</creatorcontrib><title>An Adaptive Read Control Voltage Scheme for Reliability Enhancement of Flash-Based In-Memory Computing Architecture for Neural Network</title><title>IEEE transactions on device and materials reliability</title><addtitle>TDMR</addtitle><description>The storage reliability is critical for flash memory based computing in-memory (CIM) architecture for Convolutional Neural Network (CNN). In this paper, we constructed a CIM scheme based on the Nor Flash array (NFA). We conducted simulations to investigate the impact of threshold voltage distribution and drift of Flash memory cells on the recognition accuracy for various CNN architectures based on the CIM schemes. Based on the reliability study, we proposed a novel compensation scheme to effectively mitigate the impact of threshold voltage drift and evaluated the effectiveness of the proposed scheme by recognition accuracy evaluation.</description><subject>Accuracy</subject><subject>Adaptive control</subject><subject>Artificial neural networks</subject><subject>Computation</subject><subject>Computing-in-memory</subject><subject>convolutional neural network</subject><subject>Convolutional neural networks</subject><subject>Drift</subject><subject>Flash memories</subject><subject>Flash memory (computers)</subject><subject>flash memory reliability</subject><subject>In-memory computing</subject><subject>multi-level NOR-flash</subject><subject>Network reliability</subject><subject>Recognition</subject><subject>Reliability</subject><subject>Reliability analysis</subject><subject>Threshold voltage</subject><subject>Transistors</subject><issn>1530-4388</issn><issn>1558-2574</issn><fulltext>true</fulltext><rsrctype>magazinearticle</rsrctype><creationdate>2024</creationdate><recordtype>magazinearticle</recordtype><sourceid>RIE</sourceid><recordid>eNpNkN1OgzAUx4nRxDl9ABMvmnjNbGkpcIlz0yVTkzm9JaU9DCZQLEWzF_C5hbALr85J_h8n5-c41wTPCMHR3fbheTPzsMdmlHkR596JMyG-H7qeH7DTYafYZTQMz52Ltt1jTKLA5xPnN65RrERji29AGxAKzXVtjS7Rhy6t2AF6kzlUgDJter0sRFqUhT2gRZ2LWvZKbZHO0LIUbe7eixYUWtXuM1TaHPququlsUe9QbGReWJC2M2PXC3RGlP2wP9p8XjpnmShbuDrOqfO-XGznT-769XE1j9euJAG3LlWSAxc8UBmmkYfTgECmuKRp_00kgPGMMT9SPvFlIP1UhiykAksV4RAzBXTq3I69jdFfHbQ22evO1P3JhBJMKAtDQnsXGV3S6LY1kCWNKSphDgnByYA7GXAnA-7kiLvP3IyZAgD--TkmJGL0D3J0fSc</recordid><startdate>20240901</startdate><enddate>20240901</enddate><creator>Zhang, Xinrui</creator><creator>Huang, Jian</creator><creator>Liu, Xianping</creator><creator>Zhong, Baiqing</creator><creator>Yu, Zhiyi</creator><general>IEEE</general><general>The Institute of Electrical and Electronics Engineers, Inc. (IEEE)</general><scope>97E</scope><scope>RIA</scope><scope>RIE</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SP</scope><scope>8FD</scope><scope>L7M</scope><orcidid>https://orcid.org/0009-0008-3110-5223</orcidid><orcidid>https://orcid.org/0009-0003-1106-4616</orcidid><orcidid>https://orcid.org/0009-0002-3301-6861</orcidid><orcidid>https://orcid.org/0000-0002-8802-0457</orcidid></search><sort><creationdate>20240901</creationdate><title>An Adaptive Read Control Voltage Scheme for Reliability Enhancement of Flash-Based In-Memory Computing Architecture for Neural Network</title><author>Zhang, Xinrui ; Huang, Jian ; Liu, Xianping ; Zhong, Baiqing ; Yu, Zhiyi</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c176t-3dc6e6a67df03920b71efd6c3b9759ae46f4459d515c7c5bc8483a0cd90804de3</frbrgroupid><rsrctype>magazinearticle</rsrctype><prefilter>magazinearticle</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Accuracy</topic><topic>Adaptive control</topic><topic>Artificial neural networks</topic><topic>Computation</topic><topic>Computing-in-memory</topic><topic>convolutional neural network</topic><topic>Convolutional neural networks</topic><topic>Drift</topic><topic>Flash memories</topic><topic>Flash memory (computers)</topic><topic>flash memory reliability</topic><topic>In-memory computing</topic><topic>multi-level NOR-flash</topic><topic>Network reliability</topic><topic>Recognition</topic><topic>Reliability</topic><topic>Reliability analysis</topic><topic>Threshold voltage</topic><topic>Transistors</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Zhang, Xinrui</creatorcontrib><creatorcontrib>Huang, Jian</creatorcontrib><creatorcontrib>Liu, Xianping</creatorcontrib><creatorcontrib>Zhong, Baiqing</creatorcontrib><creatorcontrib>Yu, Zhiyi</creatorcontrib><collection>IEEE All-Society Periodicals Package (ASPP) 2005-present</collection><collection>IEEE All-Society Periodicals Package (ASPP) 1998-Present</collection><collection>IEEE Xplore</collection><collection>CrossRef</collection><collection>Electronics & Communications Abstracts</collection><collection>Technology Research Database</collection><collection>Advanced Technologies Database with Aerospace</collection><jtitle>IEEE transactions on device and materials reliability</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Zhang, Xinrui</au><au>Huang, Jian</au><au>Liu, Xianping</au><au>Zhong, Baiqing</au><au>Yu, Zhiyi</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>An Adaptive Read Control Voltage Scheme for Reliability Enhancement of Flash-Based In-Memory Computing Architecture for Neural Network</atitle><jtitle>IEEE transactions on device and materials reliability</jtitle><stitle>TDMR</stitle><date>2024-09-01</date><risdate>2024</risdate><volume>24</volume><issue>3</issue><spage>422</spage><epage>427</epage><pages>422-427</pages><issn>1530-4388</issn><eissn>1558-2574</eissn><coden>ITDMA2</coden><abstract>The storage reliability is critical for flash memory based computing in-memory (CIM) architecture for Convolutional Neural Network (CNN). In this paper, we constructed a CIM scheme based on the Nor Flash array (NFA). We conducted simulations to investigate the impact of threshold voltage distribution and drift of Flash memory cells on the recognition accuracy for various CNN architectures based on the CIM schemes. Based on the reliability study, we proposed a novel compensation scheme to effectively mitigate the impact of threshold voltage drift and evaluated the effectiveness of the proposed scheme by recognition accuracy evaluation.</abstract><cop>New York</cop><pub>IEEE</pub><doi>10.1109/TDMR.2024.3429662</doi><tpages>6</tpages><orcidid>https://orcid.org/0009-0008-3110-5223</orcidid><orcidid>https://orcid.org/0009-0003-1106-4616</orcidid><orcidid>https://orcid.org/0009-0002-3301-6861</orcidid><orcidid>https://orcid.org/0000-0002-8802-0457</orcidid></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | ISSN: 1530-4388 |
| ispartof | IEEE transactions on device and materials reliability, 2024-09, Vol.24 (3), p.422-427 |
| issn | 1530-4388 1558-2574 |
| language | eng |
| recordid | cdi_ieee_primary_10601194 |
| source | IEEE Xplore |
| subjects | Accuracy Adaptive control Artificial neural networks Computation Computing-in-memory convolutional neural network Convolutional neural networks Drift Flash memories Flash memory (computers) flash memory reliability In-memory computing multi-level NOR-flash Network reliability Recognition Reliability Reliability analysis Threshold voltage Transistors |
| title | An Adaptive Read Control Voltage Scheme for Reliability Enhancement of Flash-Based In-Memory Computing Architecture for Neural Network |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-07T18%3A41%3A07IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_RIE&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=An%20Adaptive%20Read%20Control%20Voltage%20Scheme%20for%20Reliability%20Enhancement%20of%20Flash-Based%20In-Memory%20Computing%20Architecture%20for%20Neural%20Network&rft.jtitle=IEEE%20transactions%20on%20device%20and%20materials%20reliability&rft.au=Zhang,%20Xinrui&rft.date=2024-09-01&rft.volume=24&rft.issue=3&rft.spage=422&rft.epage=427&rft.pages=422-427&rft.issn=1530-4388&rft.eissn=1558-2574&rft.coden=ITDMA2&rft_id=info:doi/10.1109/TDMR.2024.3429662&rft_dat=%3Cproquest_RIE%3E3101348813%3C/proquest_RIE%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=3101348813&rft_id=info:pmid/&rft_ieee_id=10601194&rfr_iscdi=true |