Voltage-Controlled Skyrmion-Based Leaky Integrate and Fire Neurons for Spiking Neural Networks
Spintronics is an emerging technology for data storage and computation. Magnetic skyrmion-based devices are attractive in this field due to their small size and low energy consumption. However, controlling skyrmions' creation, deletion, and motion is challenging. In this article, we propose a n...
Gespeichert in:
| Veröffentlicht in: | IEEE transactions on electron devices 2024-10, Vol.71 (10), p.6395-6402 |
|---|---|
| Hauptverfasser: | , , , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 6402 |
|---|---|
| container_issue | 10 |
| container_start_page | 6395 |
| container_title | IEEE transactions on electron devices |
| container_volume | 71 |
| creator | Lone, Aijaz H. Aguirre, Fernando Rahimi, Daniel N. Lanza, Mario Setti, Gianluca Fariborzi, Hossein |
| description | Spintronics is an emerging technology for data storage and computation. Magnetic skyrmion-based devices are attractive in this field due to their small size and low energy consumption. However, controlling skyrmions' creation, deletion, and motion is challenging. In this article, we propose a novel energy-efficient skyrmion-based device structure and demonstrate its use as a leaky integrate and fire (LIF) neuron for neuromorphic computing. Using micromagnetic simulations, we show that skyrmions can be confined by patterning the geometry of the free layer (FL) in a magnetic tunnel junction (MTJ) and demonstrate that the size of the skyrmion can be adjusted by applying pulsed voltage stresses, and when tuned, the device acts as an LIF neuron. The input voltage spikes at the input terminal control the spike rate. The MTJ dissipates energy around 10.23 fJ/spike. A spiking neural network (SNN) of such skyrmion-based LIF neurons can classify images from the Modified National Institute of Standards and Technology (MNIST) dataset. |
| doi_str_mv | 10.1109/TED.2024.3442161 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_RIE</sourceid><recordid>TN_cdi_crossref_primary_10_1109_TED_2024_3442161</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>10643493</ieee_id><sourcerecordid>3109496882</sourcerecordid><originalsourceid>FETCH-LOGICAL-c175t-f8e9e6f3317ad2008152a819c55c867aaca96f7fb2e825f7f61af6dc5353814e3</originalsourceid><addsrcrecordid>eNpNkL1PwzAQxS0EEuVjZ2CIxJzi79gjlBYqVTC0MGKZ5FylSeNip0L973FpB6Z37_TenfRD6IbgISFY3y_GT0OKKR8yzimR5AQNiBBFriWXp2iAMVG5Zoqdo4sYV8nKlBugzw_f9nYJ-ch3ffBtC1U2b3ZhXfsuf7Qx2RnYZpdNux6WwfaQ2a7KJnWA7BW2wXcxcz5k803d1N3yb2fbJP2PD028QmfOthGuj3qJ3ifjxegln709T0cPs7wkhehzp0CDdIyRwlYUY0UEtYroUohSycLa0mrpCvdFQVGRBkmsk1UpmGCKcGCX6O5wdxP89xZib1Z-G7r00rBEh2upFE0pfEiVwccYwJlNqNc27AzBZk_RJIpmT9EcKabK7aFSA8C_uOSMa8Z-AYifbhA</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>3109496882</pqid></control><display><type>article</type><title>Voltage-Controlled Skyrmion-Based Leaky Integrate and Fire Neurons for Spiking Neural Networks</title><source>IEEE Electronic Library (IEL)</source><creator>Lone, Aijaz H. ; Aguirre, Fernando ; Rahimi, Daniel N. ; Lanza, Mario ; Setti, Gianluca ; Fariborzi, Hossein</creator><creatorcontrib>Lone, Aijaz H. ; Aguirre, Fernando ; Rahimi, Daniel N. ; Lanza, Mario ; Setti, Gianluca ; Fariborzi, Hossein</creatorcontrib><description>Spintronics is an emerging technology for data storage and computation. Magnetic skyrmion-based devices are attractive in this field due to their small size and low energy consumption. However, controlling skyrmions' creation, deletion, and motion is challenging. In this article, we propose a novel energy-efficient skyrmion-based device structure and demonstrate its use as a leaky integrate and fire (LIF) neuron for neuromorphic computing. Using micromagnetic simulations, we show that skyrmions can be confined by patterning the geometry of the free layer (FL) in a magnetic tunnel junction (MTJ) and demonstrate that the size of the skyrmion can be adjusted by applying pulsed voltage stresses, and when tuned, the device acts as an LIF neuron. The input voltage spikes at the input terminal control the spike rate. The MTJ dissipates energy around 10.23 fJ/spike. A spiking neural network (SNN) of such skyrmion-based LIF neurons can classify images from the Modified National Institute of Standards and Technology (MNIST) dataset.</description><identifier>ISSN: 0018-9383</identifier><identifier>EISSN: 1557-9646</identifier><identifier>DOI: 10.1109/TED.2024.3442161</identifier><identifier>CODEN: IETDAI</identifier><language>eng</language><publisher>New York: IEEE</publisher><subject>Anisotropic magnetoresistance ; Data storage ; Electric potential ; Energy consumption ; Hypothetical particles ; Leaky integrate and fire (LIF) neuron ; Magnetic confinement ; magnetic tunnel junction (MTJ) ; Magnetic tunneling ; Magnetization ; Mathematical models ; Neural networks ; neuromorphic computing ; Neurons ; Particle theory ; skyrmion ; Skyrmions ; spiking neural network (SNN) ; Spintronics ; Tunnel junctions ; Voltage</subject><ispartof>IEEE transactions on electron devices, 2024-10, Vol.71 (10), p.6395-6402</ispartof><rights>Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2024</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c175t-f8e9e6f3317ad2008152a819c55c867aaca96f7fb2e825f7f61af6dc5353814e3</cites><orcidid>0000-0002-1687-2917 ; 0000-0002-7828-0239 ; 0000-0002-2496-1856 ; 0000-0003-4756-8632</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/10643493$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>314,776,780,792,27901,27902,54733</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/10643493$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Lone, Aijaz H.</creatorcontrib><creatorcontrib>Aguirre, Fernando</creatorcontrib><creatorcontrib>Rahimi, Daniel N.</creatorcontrib><creatorcontrib>Lanza, Mario</creatorcontrib><creatorcontrib>Setti, Gianluca</creatorcontrib><creatorcontrib>Fariborzi, Hossein</creatorcontrib><title>Voltage-Controlled Skyrmion-Based Leaky Integrate and Fire Neurons for Spiking Neural Networks</title><title>IEEE transactions on electron devices</title><addtitle>TED</addtitle><description>Spintronics is an emerging technology for data storage and computation. Magnetic skyrmion-based devices are attractive in this field due to their small size and low energy consumption. However, controlling skyrmions' creation, deletion, and motion is challenging. In this article, we propose a novel energy-efficient skyrmion-based device structure and demonstrate its use as a leaky integrate and fire (LIF) neuron for neuromorphic computing. Using micromagnetic simulations, we show that skyrmions can be confined by patterning the geometry of the free layer (FL) in a magnetic tunnel junction (MTJ) and demonstrate that the size of the skyrmion can be adjusted by applying pulsed voltage stresses, and when tuned, the device acts as an LIF neuron. The input voltage spikes at the input terminal control the spike rate. The MTJ dissipates energy around 10.23 fJ/spike. A spiking neural network (SNN) of such skyrmion-based LIF neurons can classify images from the Modified National Institute of Standards and Technology (MNIST) dataset.</description><subject>Anisotropic magnetoresistance</subject><subject>Data storage</subject><subject>Electric potential</subject><subject>Energy consumption</subject><subject>Hypothetical particles</subject><subject>Leaky integrate and fire (LIF) neuron</subject><subject>Magnetic confinement</subject><subject>magnetic tunnel junction (MTJ)</subject><subject>Magnetic tunneling</subject><subject>Magnetization</subject><subject>Mathematical models</subject><subject>Neural networks</subject><subject>neuromorphic computing</subject><subject>Neurons</subject><subject>Particle theory</subject><subject>skyrmion</subject><subject>Skyrmions</subject><subject>spiking neural network (SNN)</subject><subject>Spintronics</subject><subject>Tunnel junctions</subject><subject>Voltage</subject><issn>0018-9383</issn><issn>1557-9646</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>RIE</sourceid><recordid>eNpNkL1PwzAQxS0EEuVjZ2CIxJzi79gjlBYqVTC0MGKZ5FylSeNip0L973FpB6Z37_TenfRD6IbgISFY3y_GT0OKKR8yzimR5AQNiBBFriWXp2iAMVG5Zoqdo4sYV8nKlBugzw_f9nYJ-ch3ffBtC1U2b3ZhXfsuf7Qx2RnYZpdNux6WwfaQ2a7KJnWA7BW2wXcxcz5k803d1N3yb2fbJP2PD028QmfOthGuj3qJ3ifjxegln709T0cPs7wkhehzp0CDdIyRwlYUY0UEtYroUohSycLa0mrpCvdFQVGRBkmsk1UpmGCKcGCX6O5wdxP89xZib1Z-G7r00rBEh2upFE0pfEiVwccYwJlNqNc27AzBZk_RJIpmT9EcKabK7aFSA8C_uOSMa8Z-AYifbhA</recordid><startdate>20241001</startdate><enddate>20241001</enddate><creator>Lone, Aijaz H.</creator><creator>Aguirre, Fernando</creator><creator>Rahimi, Daniel N.</creator><creator>Lanza, Mario</creator><creator>Setti, Gianluca</creator><creator>Fariborzi, Hossein</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/0000-0002-1687-2917</orcidid><orcidid>https://orcid.org/0000-0002-7828-0239</orcidid><orcidid>https://orcid.org/0000-0002-2496-1856</orcidid><orcidid>https://orcid.org/0000-0003-4756-8632</orcidid></search><sort><creationdate>20241001</creationdate><title>Voltage-Controlled Skyrmion-Based Leaky Integrate and Fire Neurons for Spiking Neural Networks</title><author>Lone, Aijaz H. ; Aguirre, Fernando ; Rahimi, Daniel N. ; Lanza, Mario ; Setti, Gianluca ; Fariborzi, Hossein</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c175t-f8e9e6f3317ad2008152a819c55c867aaca96f7fb2e825f7f61af6dc5353814e3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Anisotropic magnetoresistance</topic><topic>Data storage</topic><topic>Electric potential</topic><topic>Energy consumption</topic><topic>Hypothetical particles</topic><topic>Leaky integrate and fire (LIF) neuron</topic><topic>Magnetic confinement</topic><topic>magnetic tunnel junction (MTJ)</topic><topic>Magnetic tunneling</topic><topic>Magnetization</topic><topic>Mathematical models</topic><topic>Neural networks</topic><topic>neuromorphic computing</topic><topic>Neurons</topic><topic>Particle theory</topic><topic>skyrmion</topic><topic>Skyrmions</topic><topic>spiking neural network (SNN)</topic><topic>Spintronics</topic><topic>Tunnel junctions</topic><topic>Voltage</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Lone, Aijaz H.</creatorcontrib><creatorcontrib>Aguirre, Fernando</creatorcontrib><creatorcontrib>Rahimi, Daniel N.</creatorcontrib><creatorcontrib>Lanza, Mario</creatorcontrib><creatorcontrib>Setti, Gianluca</creatorcontrib><creatorcontrib>Fariborzi, Hossein</creatorcontrib><collection>IEEE All-Society Periodicals Package (ASPP) 2005-present</collection><collection>IEEE All-Society Periodicals Package (ASPP) 1998-Present</collection><collection>IEEE Electronic Library (IEL)</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 electron devices</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Lone, Aijaz H.</au><au>Aguirre, Fernando</au><au>Rahimi, Daniel N.</au><au>Lanza, Mario</au><au>Setti, Gianluca</au><au>Fariborzi, Hossein</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Voltage-Controlled Skyrmion-Based Leaky Integrate and Fire Neurons for Spiking Neural Networks</atitle><jtitle>IEEE transactions on electron devices</jtitle><stitle>TED</stitle><date>2024-10-01</date><risdate>2024</risdate><volume>71</volume><issue>10</issue><spage>6395</spage><epage>6402</epage><pages>6395-6402</pages><issn>0018-9383</issn><eissn>1557-9646</eissn><coden>IETDAI</coden><abstract>Spintronics is an emerging technology for data storage and computation. Magnetic skyrmion-based devices are attractive in this field due to their small size and low energy consumption. However, controlling skyrmions' creation, deletion, and motion is challenging. In this article, we propose a novel energy-efficient skyrmion-based device structure and demonstrate its use as a leaky integrate and fire (LIF) neuron for neuromorphic computing. Using micromagnetic simulations, we show that skyrmions can be confined by patterning the geometry of the free layer (FL) in a magnetic tunnel junction (MTJ) and demonstrate that the size of the skyrmion can be adjusted by applying pulsed voltage stresses, and when tuned, the device acts as an LIF neuron. The input voltage spikes at the input terminal control the spike rate. The MTJ dissipates energy around 10.23 fJ/spike. A spiking neural network (SNN) of such skyrmion-based LIF neurons can classify images from the Modified National Institute of Standards and Technology (MNIST) dataset.</abstract><cop>New York</cop><pub>IEEE</pub><doi>10.1109/TED.2024.3442161</doi><tpages>8</tpages><orcidid>https://orcid.org/0000-0002-1687-2917</orcidid><orcidid>https://orcid.org/0000-0002-7828-0239</orcidid><orcidid>https://orcid.org/0000-0002-2496-1856</orcidid><orcidid>https://orcid.org/0000-0003-4756-8632</orcidid></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | ISSN: 0018-9383 |
| ispartof | IEEE transactions on electron devices, 2024-10, Vol.71 (10), p.6395-6402 |
| issn | 0018-9383 1557-9646 |
| language | eng |
| recordid | cdi_crossref_primary_10_1109_TED_2024_3442161 |
| source | IEEE Electronic Library (IEL) |
| subjects | Anisotropic magnetoresistance Data storage Electric potential Energy consumption Hypothetical particles Leaky integrate and fire (LIF) neuron Magnetic confinement magnetic tunnel junction (MTJ) Magnetic tunneling Magnetization Mathematical models Neural networks neuromorphic computing Neurons Particle theory skyrmion Skyrmions spiking neural network (SNN) Spintronics Tunnel junctions Voltage |
| title | Voltage-Controlled Skyrmion-Based Leaky Integrate and Fire Neurons for Spiking Neural Networks |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-08T17%3A23%3A25IST&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=Voltage-Controlled%20Skyrmion-Based%20Leaky%20Integrate%20and%20Fire%20Neurons%20for%20Spiking%20Neural%20Networks&rft.jtitle=IEEE%20transactions%20on%20electron%20devices&rft.au=Lone,%20Aijaz%20H.&rft.date=2024-10-01&rft.volume=71&rft.issue=10&rft.spage=6395&rft.epage=6402&rft.pages=6395-6402&rft.issn=0018-9383&rft.eissn=1557-9646&rft.coden=IETDAI&rft_id=info:doi/10.1109/TED.2024.3442161&rft_dat=%3Cproquest_RIE%3E3109496882%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=3109496882&rft_id=info:pmid/&rft_ieee_id=10643493&rfr_iscdi=true |