Reliable Counterfactual Semantic Transmission for Energy-Efficient Visual IoT in Intelligent Environments | 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 Reliable Counterfactual Semantic Transmission for Energy-Efficient Visual IoT in Intelligent Environments David Elias Emmanuel, Anil Fernando This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9542617/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 4 You are reading this latest preprint version Abstract Visual Internet of Things (VIoT) systems are increasingly used in intelligent environments for traffic monitoring, intrusion detection, and scene-aware surveillance. However, many deployments still follow a transmit-first paradigm, where video is continuously streamed even when much of the visual data does not alter the downstream decision. This creates unnecessary bitrate, energy, and latency overhead for resource-constrained edge devices.This paper presents Counterfactual Semantic Transmission for Visual IoT (CST-VIoT) , a decision-aware framework that transmits only the minimal semantic evidence required to alter or preserve the receiver's decision. CST-VIoT integrates object detection, scene-graph construction, lightweight decision modelling, counterfactual perturbation search, and semantic packet encoding into a unified edge-to-receiver pipeline. Evaluated across six surveillance-oriented scenarios derived from UA-DETRAC and CDNet, CST-VIoT achieves a mean decision F1 score of \((0.901 \pm 0.018)\), while reducing bitrate, energy consumption, and latency by \((96.9%)\), \((86.7%)\), and \((70.4%)\), respectively, relative to H.265 transmission, with only a \((2.2%)\) relative reduction in F1. Ablation, sensitivity, and robustness analyses further show graceful degradation under frame noise and packet loss. These results indicate that counterfactual, decision-aware communication can support reliable and energy-efficient VIoT operation in intelligent environments under bandwidth and energy constraints. Visual Internet of Things intelligent environments semantic communication counterfactual reasoning edge intelligence energy-efficient transmission scene graphs Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers invited by journal 19 May, 2026 Editor assigned by journal 12 May, 2026 Submission checks completed at journal 28 Apr, 2026 First submitted to journal 27 Apr, 2026 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-9542617","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":631430421,"identity":"45fa72a4-d905-4a53-ae5e-2268e38db55f","order_by":0,"name":"David Elias Emmanuel","email":"data:image/png;base64,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","orcid":"","institution":"University of Strathclyde","correspondingAuthor":true,"prefix":"","firstName":"David","middleName":"Elias","lastName":"Emmanuel","suffix":""},{"id":631430424,"identity":"8d5faad8-2a8f-4f01-9d55-fb2275ac30b3","order_by":1,"name":"Anil Fernando","email":"","orcid":"","institution":"University of Strathclyde","correspondingAuthor":false,"prefix":"","firstName":"Anil","middleName":"","lastName":"Fernando","suffix":""}],"badges":[],"createdAt":"2026-04-27 13:40:29","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9542617/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9542617/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108806027,"identity":"b232b4db-0ac2-487d-8db9-d8d4ba502f35","added_by":"auto","created_at":"2026-05-08 15:27:31","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1538254,"visible":true,"origin":"","legend":"","description":"","filename":"CSTVIoTforIntelligentEnvironments.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9542617/v1_covered_0e1a0d31-ffbe-49a1-b9ea-f57413f56d9c.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Reliable Counterfactual Semantic Transmission for Energy-Efficient Visual IoT in Intelligent Environments","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"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":"journal-of-reliable-intelligent-environments","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jrie","sideBox":"Learn more about [Journal of Reliable Intelligent Environments](http://link.springer.com/journal/40860)","snPcode":"40860","submissionUrl":"https://submission.nature.com/new-submission/40860/3","title":"Journal of Reliable Intelligent Environments","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Visual Internet of Things, intelligent environments, semantic communication, counterfactual reasoning, edge intelligence, energy-efficient transmission, scene graphs","lastPublishedDoi":"10.21203/rs.3.rs-9542617/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9542617/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eVisual Internet of Things (VIoT) systems are increasingly used in intelligent environments for traffic monitoring, intrusion detection, and scene-aware surveillance. However, many deployments still follow a transmit-first paradigm, where video is continuously streamed even when much of the visual data does not alter the downstream decision.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThis creates unnecessary bitrate, energy, and latency overhead for resource-constrained edge devices.This paper presents \u003cstrong\u003eCounterfactual Semantic Transmission for Visual IoT (CST-VIoT)\u003c/strong\u003e, a decision-aware framework that transmits only the minimal semantic evidence required to alter or preserve the receiver's decision. CST-VIoT integrates object detection, scene-graph construction, lightweight decision modelling, counterfactual perturbation search, and semantic packet encoding into a unified edge-to-receiver pipeline.\u003c/p\u003e\n\u003cp\u003eEvaluated across six surveillance-oriented scenarios derived from UA-DETRAC and CDNet, CST-VIoT achieves a mean decision F1 score of \\((0.901 \\pm 0.018)\\), while reducing bitrate, energy consumption, and latency by \\((96.9%)\\), \\((86.7%)\\), and \\((70.4%)\\), respectively, relative to H.265 transmission, with only a \\((2.2%)\\) relative reduction in F1. Ablation, sensitivity, and robustness analyses further show graceful degradation under frame noise and packet loss.\u003c/p\u003e\n\u003cp\u003eThese results indicate that counterfactual, decision-aware communication can support reliable and energy-efficient VIoT operation in intelligent environments under bandwidth and energy constraints.\u003c/p\u003e","manuscriptTitle":"Reliable Counterfactual Semantic Transmission for Energy-Efficient Visual IoT in Intelligent Environments","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-07 04:59:03","doi":"10.21203/rs.3.rs-9542617/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewersInvited","content":"","date":"2026-05-19T10:58:21+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-05-12T08:50:24+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-29T03:31:09+00:00","index":"","fulltext":""},{"type":"submitted","content":"Journal of Reliable Intelligent Environments","date":"2026-04-27T13:23:43+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"
[email protected]","identity":"journal-of-reliable-intelligent-environments","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jrie","sideBox":"Learn more about [Journal of Reliable Intelligent Environments](http://link.springer.com/journal/40860)","snPcode":"40860","submissionUrl":"https://submission.nature.com/new-submission/40860/3","title":"Journal of Reliable Intelligent Environments","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"61a934b7-4618-4a5f-8035-8bbd85916d0e","owner":[],"postedDate":"May 7th, 2026","published":true,"recentEditorialEvents":[{"type":"reviewersInvited","content":"2","date":"2026-05-19T10:58:21+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-05-12T08:50:24+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-19T11:08:25+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-07 04:59:03","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9542617","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9542617","identity":"rs-9542617","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}
Text is read by the "Ask this paper" AI Q&A widget below.
Extraction quality varies by source — PMC NXML preserves structure
cleanly, OA-HTML may include some navigation residue, and OA-PDF can
have broken hyphenation. The publisher copy
(via DOI)
is the canonical version.