Deploying Human Activity Recognition in Embedded RISC-V Processors

Deploying Human Activity Recognition in Embedded RISC-V Processors | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Deploying Human Activity Recognition in Embedded RISC-V Processors Willian Analdo Nunes, Rafael Schild Reusch, Lucas Luza, Angelo Elias Dal Zotto, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4619217/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 16 Nov, 2024 Read the published version in Design Automation for Embedded Systems → Version 1 posted 10 You are reading this latest preprint version Abstract Human Activity Recognition (HAR) is an important area of research due to its applications in health monitoring, elderly care, and personal fitness tracking. The challenge is deploying efficient and accurate HAR systems on resource-constrained embedded devices, which require low power consumption and processing efficiency. This work optimizes a Convolutional Neural Network (CNN) model for HAR, targeting resource-constrained processors. The goal is to balance accuracy, performance, and power consumption for real-world deployment in wearable devices. Key contributions include introducing an Extended 1D CNN model that enhances temporal awareness and accuracy without the overhead of floating-point computations, evaluating and applying quantization methods to minimize model size with minimal accuracy loss, and assessing the model's performance on a RISC-V processor. Results show an accuracy increase from 74% to 87.2%. Memory optimization using Lookup Table (LUT) quantization reduces the memory required for model parameters by 57%. This research underscores the potential for advanced neural network models on low-power RISC-V processors in real-time HAR, with significant implications for health monitoring and smart environments. Machine Learning 1D CNN Human Activity Recognition RISC-V Embedded Systems Constrained Devices Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 16 Nov, 2024 Read the published version in Design Automation for Embedded Systems → Version 1 posted Editorial decision: Revision requested 01 Oct, 2024 Reviews received at journal 30 Jul, 2024 Reviews received at journal 11 Jul, 2024 Reviewers agreed at journal 11 Jul, 2024 Reviewers agreed at journal 10 Jul, 2024 Reviewers agreed at journal 09 Jul, 2024 Reviewers invited by journal 09 Jul, 2024 Editor assigned by journal 05 Jul, 2024 Submission checks completed at journal 24 Jun, 2024 First submitted to journal 21 Jun, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4619217","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":325712168,"identity":"371190f7-fda4-4489-b4f2-1bbe8322f4fa","order_by":0,"name":"Willian Analdo Nunes","email":"","orcid":"","institution":"Pontifical Catholic University of Rio Grande do Sul","correspondingAuthor":false,"prefix":"","firstName":"Willian","middleName":"Analdo","lastName":"Nunes","suffix":""},{"id":325712172,"identity":"fab68621-32f1-42ad-8d1e-b8d79e7d90ae","order_by":1,"name":"Rafael Schild Reusch","email":"","orcid":"","institution":"Pontifical Catholic University of Rio Grande do Sul","correspondingAuthor":false,"prefix":"","firstName":"Rafael","middleName":"Schild","lastName":"Reusch","suffix":""},{"id":325712173,"identity":"f66f3409-64fe-4092-b161-58b4bfbc2fb7","order_by":2,"name":"Lucas Luza","email":"","orcid":"","institution":"Pontifical Catholic University of Rio Grande do Sul","correspondingAuthor":false,"prefix":"","firstName":"Lucas","middleName":"","lastName":"Luza","suffix":""},{"id":325712174,"identity":"d61ab04c-90fc-412c-bc42-3119aa3196ba","order_by":3,"name":"Angelo Elias Dal Zotto","email":"","orcid":"","institution":"Pontifical Catholic University of Rio Grande do Sul","correspondingAuthor":false,"prefix":"","firstName":"Angelo","middleName":"Elias Dal","lastName":"Zotto","suffix":""},{"id":325712175,"identity":"39cd1692-533b-4bd8-b3ad-2f28d1a7f09a","order_by":4,"name":"Leonardo Rezende Juracy","email":"","orcid":"","institution":"Pontifical Catholic University of Rio Grande do Sul","correspondingAuthor":false,"prefix":"","firstName":"Leonardo","middleName":"Rezende","lastName":"Juracy","suffix":""},{"id":325712176,"identity":"c75fc745-3f0b-4be6-b67b-ca77fbf7cfda","order_by":5,"name":"Fernando Gehm Moraes","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8klEQVRIiWNgGAWjYBAC9gYwxQwXkIPSB3Bq4TmApsWYdC2JDQS1sPc+e/Bzh7Ucg9jhYx9+VNSlb5dIfvbxB8OdfJxaeI6bG/aeSTdmkE5Lntlz5nDuzhlpxrN5GJ5ZNuDQYi+RxibB23Y4sUE6x5iBt+1A7oYbOcxAdx42wGmL/DM2yb9QLYx//9WlGwC1MP7Ap0WCjU0aZgszbwNzAkgLAw8+LTxpbNKybenGbEC/MMscO2y4s+eZMTOPwTPcWtiPsUm+bbOW45dOPsz4pqZO3pw9+THjj4o7OLXAARuMYYBEEglIUjwKRsEoGAUjAgAAv5tNIM3h7McAAAAASUVORK5CYII=","orcid":"","institution":"Pontifical Catholic University of Rio Grande do Sul","correspondingAuthor":true,"prefix":"","firstName":"Fernando","middleName":"Gehm","lastName":"Moraes","suffix":""}],"badges":[],"createdAt":"2024-06-21 20:25:05","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4619217/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4619217/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s10617-024-09288-w","type":"published","date":"2024-11-16T15:57:02+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":69274806,"identity":"50604138-0fd7-48ef-88c0-4d29fee86b65","added_by":"auto","created_at":"2024-11-18 16:31:02","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":790024,"visible":true,"origin":"","legend":"","description":"","filename":"WILLIANDAES.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4619217/v1_covered_5f71f423-0c5c-4682-8429-15107b31f7b5.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Deploying Human Activity Recognition in Embedded RISC-V Processors","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"design-automation-for-embedded-systems","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"daem","sideBox":"Learn more about [Design Automation for Embedded Systems](http://link.springer.com/journal/10617)","snPcode":"10617","submissionUrl":"https://submission.nature.com/new-submission/10617/3","title":"Design Automation for Embedded Systems","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Machine Learning, 1D CNN, Human Activity Recognition, RISC-V, Embedded Systems, Constrained Devices","lastPublishedDoi":"10.21203/rs.3.rs-4619217/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4619217/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Human Activity Recognition (HAR) is an important area of research due to its applications in health monitoring, elderly care, and personal fitness tracking. The challenge is deploying efficient and accurate HAR systems on resource-constrained embedded devices, which require low power consumption and processing efficiency. This work optimizes a Convolutional Neural Network (CNN) model for HAR, targeting resource-constrained processors. The goal is to balance accuracy, performance, and power consumption for real-world deployment in wearable devices. Key contributions include introducing an Extended 1D CNN model that enhances temporal awareness and accuracy without the overhead of floating-point computations, evaluating and applying quantization methods to minimize model size with minimal accuracy loss, and assessing the model's performance on a RISC-V processor. Results show an accuracy increase from 74% to 87.2%. Memory optimization using Lookup Table (LUT) quantization reduces the memory required for model parameters by 57%. This research underscores the potential for advanced neural network models on low-power RISC-V processors in real-time HAR, with significant implications for health monitoring and smart environments.","manuscriptTitle":"Deploying Human Activity Recognition in Embedded RISC-V Processors","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-07-16 17:06:24","doi":"10.21203/rs.3.rs-4619217/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-10-01T12:49:49+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-07-30T22:50:44+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-07-11T14:10:54+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"286570181152325475875652269804088272463","date":"2024-07-11T12:07:50+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"24818116550629465002261822490019337792","date":"2024-07-10T07:57:06+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"152425263231681739485014492668864233655","date":"2024-07-09T18:59:37+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-07-09T14:35:59+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-07-05T12:52:58+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-06-24T11:03:38+00:00","index":"","fulltext":""},{"type":"submitted","content":"Design Automation for Embedded Systems","date":"2024-06-21T20:23:54+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"design-automation-for-embedded-systems","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"daem","sideBox":"Learn more about [Design Automation for Embedded Systems](http://link.springer.com/journal/10617)","snPcode":"10617","submissionUrl":"https://submission.nature.com/new-submission/10617/3","title":"Design Automation for Embedded Systems","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"4fc12c4f-2ff6-4a15-9907-4a302b94b89b","owner":[],"postedDate":"July 16th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2024-11-18T15:59:46+00:00","versionOfRecord":{"articleIdentity":"rs-4619217","link":"https://doi.org/10.1007/s10617-024-09288-w","journal":{"identity":"design-automation-for-embedded-systems","isVorOnly":false,"title":"Design Automation for Embedded Systems"},"publishedOn":"2024-11-16 15:57:02","publishedOnDateReadable":"November 16th, 2024"},"versionCreatedAt":"2024-07-16 17:06:24","video":"","vorDoi":"10.1007/s10617-024-09288-w","vorDoiUrl":"https://doi.org/10.1007/s10617-024-09288-w","workflowStages":[]},"version":"v1","identity":"rs-4619217","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4619217","identity":"rs-4619217","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}