Efficient Uncertainty Aware Human Activity Recognition On Microcontrollers Using Hyperdimensional Computing And Conformal Prediction

Efficient Uncertainty Aware Human Activity Recognition On Microcontrollers Using Hyperdimensional Computing And Conformal Prediction | 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 Article Efficient Uncertainty Aware Human Activity Recognition On Microcontrollers Using Hyperdimensional Computing And Conformal Prediction Ismail Lamaakal, Chaymae Yahyati, Yassine Maleh, Khalid El Makkaoui, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7728157/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract In this paper, we tackle a core challenge for wearable human activity recognition (HAR): delivering reliable, interpretable uncertainty on tiny microcontrollers where latency, RAM, and energy are tightly constrained. Existing embedded approaches either calibrate softmax confidences—cheap but brittle under sensor placement or tempo shifts—or rely on generative or posterior-sampling schemes that exceed TinyML budgets. We propose HDUQ-HAR, an on-device, hyperdimensional (HDC) framework that encodes each IMU window into a bipolar hypervector, classifies via prototype similarity, and derives three complementary, lightweight uncertainty signals from the same representation: (i) distance to the class prototype, (ii) similarity gap to the runner-up, and (iii) vote-dispersion capturing n-gram consensus. A label-conditional conformal layer converts these scores into set-valued predictions with finite-sample coverage guarantees and exposes a human-readable reason code indicating why uncertainty increased. Across UCI HAR, WISDM, PAMAP2, and OPPORTUNITY with subject-out splits and realistic shifts (orientation, gain, time-warp, missing axis, cross-placement), HDUQ-HAR achieves near-target coverage at 1−δ = 0.90 with near-singleton sets on i.i.d. data (average size 1.18–1.25) and robust shift/OOD detection (AUROC 0.92–0.96), while running in 3–5 ms/window on Cortex-M4 with ∼6–9KB RAM and ∼5–7KB Flash. By unifying HDC geometry with label-conditional conformal prediction, our method shows that efficiency and reliability can co-exist in wearables—yielding small, calibrated sets that expand gracefully under shift and actionable explanations practitioners can trust. Physical sciences/Engineering Physical sciences/Mathematics and computing Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted 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-7728157","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":537461594,"identity":"42173858-d3b7-491b-b8ee-b535d8695d8a","order_by":0,"name":"Ismail Lamaakal","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABBklEQVRIiWNgGAWjYBACNhCRUMAgA6IZGxgs5NiI02LAwAPVImFMUAsEIGlJbCCkmI//dOKHB0At/NMOP/w4o0YivY+9+eADhop7drj0sknkbpYAOUzidpqx5IZjErltPMeSDRjOFCfj1sK7AayF4XaCgeQDoAltEjlmEoxtCcm4HMbGf3bzD5AW+dvpn38++CeRzib//vsPvFoYcreBbTG4nWMmubFNIoFNgoeNAajFDqcWidxtFgkGEjyGt3PKLGf2SRi28aQZSyScSUjApUW+/+zmmz8qbOTkbqdvvtnzzUZevv3www8fKhLscWmBAgk0PtAKwhGEAQjZMgpGwSgYBSMHAAAPck/VOkwKZwAAAABJRU5ErkJggg==","orcid":"","institution":"Mohammed Premier University","correspondingAuthor":true,"prefix":"","firstName":"Ismail","middleName":"","lastName":"Lamaakal","suffix":""},{"id":537461595,"identity":"acdeda5f-2288-42f9-9f99-da7cdfe1d632","order_by":1,"name":"Chaymae Yahyati","email":"","orcid":"","institution":"Mohammed Premier University","correspondingAuthor":false,"prefix":"","firstName":"Chaymae","middleName":"","lastName":"Yahyati","suffix":""},{"id":537461596,"identity":"1ce59e0b-a886-4859-a87d-fc771552b2c9","order_by":2,"name":"Yassine Maleh","email":"","orcid":"","institution":"Sultan Moulay Slimane University","correspondingAuthor":false,"prefix":"","firstName":"Yassine","middleName":"","lastName":"Maleh","suffix":""},{"id":537461597,"identity":"cd3a8800-b668-4ead-81d8-44f9eb41ce2a","order_by":3,"name":"Khalid El Makkaoui","email":"","orcid":"","institution":"Mohammed Premier University","correspondingAuthor":false,"prefix":"","firstName":"Khalid","middleName":"El","lastName":"Makkaoui","suffix":""},{"id":537461598,"identity":"56c4e666-58cb-4ea4-a3ee-8f7367b94b06","order_by":4,"name":"Ibrahim Ouahbi","email":"","orcid":"","institution":"Mohammed Premier University","correspondingAuthor":false,"prefix":"","firstName":"Ibrahim","middleName":"","lastName":"Ouahbi","suffix":""},{"id":537461599,"identity":"f0cc9970-dd55-4ad7-80a6-e544a4b72e44","order_by":5,"name":"Ahmad Alanezi","email":"","orcid":"","institution":"Saad Alabdullah police academy","correspondingAuthor":false,"prefix":"","firstName":"Ahmad","middleName":"","lastName":"Alanezi","suffix":""},{"id":537461600,"identity":"8a6adfd7-1a29-4fb9-9db1-d78b64b50163","order_by":6,"name":"Ahmed A. Abd EL-Latif","email":"","orcid":"","institution":"Prince Sultan University","correspondingAuthor":false,"prefix":"","firstName":"Ahmed","middleName":"A. Abd","lastName":"EL-Latif","suffix":""}],"badges":[],"createdAt":"2025-09-27 11:38:15","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7728157/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7728157/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":94863423,"identity":"579eef33-6eb2-4015-974c-038ac6858738","added_by":"auto","created_at":"2025-10-31 13:32:01","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":9471,"visible":true,"origin":"","legend":"","description":"","filename":"ca44bf954d3d47c18a981c83d779195a.json","url":"https://assets-eu.researchsquare.com/files/rs-7728157/v1/4331841fc15de275ef75d614.json"},{"id":101397890,"identity":"5388c88d-7adf-4200-838f-fd16484ebf0a","added_by":"auto","created_at":"2026-01-29 09:37:57","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1338072,"visible":true,"origin":"","legend":"","description":"","filename":"ScientificReportIsmailPaper9PhDthesisVersion2withouthighlights.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7728157/v1_covered_afc18508-a113-4a60-894b-31666750cd5a.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Efficient Uncertainty Aware Human Activity Recognition On Microcontrollers Using Hyperdimensional Computing And Conformal Prediction","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7728157/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7728157/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIn this paper, we tackle a core challenge for wearable human activity recognition (HAR): delivering reliable, interpretable uncertainty on tiny microcontrollers where latency, RAM, and energy are tightly constrained. Existing embedded approaches either calibrate softmax confidences—cheap but brittle under sensor placement or tempo shifts—or rely on generative or posterior-sampling schemes that exceed TinyML budgets. We propose HDUQ-HAR, an on-device, hyperdimensional (HDC) framework that encodes each IMU window into a bipolar hypervector, classifies via prototype similarity, and derives three complementary, lightweight uncertainty signals from the same representation: (i) distance to the class prototype, (ii) similarity gap to the runner-up, and (iii) vote-dispersion capturing n-gram consensus. A label-conditional conformal layer converts these scores into set-valued predictions with finite-sample coverage guarantees and exposes a human-readable reason code indicating why uncertainty increased. Across UCI HAR, WISDM, PAMAP2, and OPPORTUNITY with subject-out splits and realistic shifts (orientation, gain, time-warp, missing axis, cross-placement), HDUQ-HAR achieves near-target coverage at 1−δ = 0.90 with near-singleton sets on i.i.d. data (average size 1.18–1.25) and robust shift/OOD detection (AUROC 0.92–0.96), while running in 3–5 ms/window on Cortex-M4 with ∼6–9KB RAM and ∼5–7KB Flash. By unifying HDC geometry with label-conditional conformal prediction, our method shows that efficiency and reliability can co-exist in wearables—yielding small, calibrated sets that expand gracefully under shift and actionable explanations practitioners can trust.\u003c/p\u003e","manuscriptTitle":"Efficient Uncertainty Aware Human Activity Recognition On Microcontrollers Using Hyperdimensional Computing And Conformal Prediction","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-31 13:31:57","doi":"10.21203/rs.3.rs-7728157/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"3b1fed1d-c770-44b5-8d22-54ea637d304b","owner":[],"postedDate":"October 31st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":57158302,"name":"Physical sciences/Engineering"},{"id":57158303,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2026-01-28T10:08:07+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-31 13:31:57","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7728157","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7728157","identity":"rs-7728157","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}