EcoFedX- Adaptive Multi-Objective Federated Learning for Energy- Efficient Image and Speech Signal Processing at the Edge 

EcoFedX- Adaptive Multi-Objective Federated Learning for Energy- Efficient Image and Speech Signal Processing at the Edge | 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 EcoFedX- Adaptive Multi-Objective Federated Learning for Energy- Efficient Image and Speech Signal Processing at the Edge Charu Gupta, Nitasha Rathore, Gargi Mishra This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7818007/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 Federated learning (FL) enables distributed model training across edge devices while preserving data privacy, yet existing methods struggle to balance accuracy, communication cost, and energy efficiency under heterogeneous conditions. This paper introduces EcoFedX, an adaptive multi-objective FL framework that jointly optimizes these three objectives through entropy- and energy-aware weighting of local updates. EcoFedX dynamically adjusts client contributions based on data diversity and device power profiles, achieving both statistical robustness and system-level sustainability. Extensive experiments on CIFAR-10, Google Speech Commands, and UCI-HAR datasets demonstrate that EcoFedX achieves ≈4 % higher accuracy, ≈38 ± 2 % lower communication cost, and ≈30 ± 3 % lower energy consumption relative to FedAvg, without compromising convergence. Ablation and sensitivity analyses further validate the framework’s stability under varying compression ratios, client heterogeneity, and energy budgets (Supplementary S7). The results confirm that EcoFedX offers a principled, interpretable, and resource-efficient foundation for scalable edge intelligence. Federated Learning Multi-Objective Optimization Edge AI Signal Processing Image and Speech Recognition Energy Efficiency TinyML Adaptive Optimization Full Text Additional Declarations No competing interests reported. Supplementary Files suppletaryfedexfinal.docx Cite Share Download PDF Status: Under Review Version 1 posted Reviewers invited by journal 12 Oct, 2025 Editor assigned by journal 10 Oct, 2025 Submission checks completed at journal 10 Oct, 2025 First submitted to journal 09 Oct, 2025 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-7818007","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":533129600,"identity":"f0c05473-ecfe-4c72-bd9e-1693ae24462f","order_by":0,"name":"Charu Gupta","email":"data:image/png;base64,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","orcid":"","institution":"Independent Researcher","correspondingAuthor":true,"prefix":"","firstName":"Charu","middleName":"","lastName":"Gupta","suffix":""},{"id":533129604,"identity":"e6b2e138-509b-49f5-ba57-0498e2d8b68b","order_by":1,"name":"Nitasha Rathore","email":"","orcid":"","institution":"Bharati Vidyapeeth Deemed University","correspondingAuthor":false,"prefix":"","firstName":"Nitasha","middleName":"","lastName":"Rathore","suffix":""},{"id":533129607,"identity":"409d3db2-b772-4cb2-b8f3-2ca43e952607","order_by":2,"name":"Gargi Mishra","email":"","orcid":"","institution":"Bharati Vidyapeeth Deemed University","correspondingAuthor":false,"prefix":"","firstName":"Gargi","middleName":"","lastName":"Mishra","suffix":""}],"badges":[],"createdAt":"2025-10-09 13:23:25","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7818007/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7818007/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":94376194,"identity":"8c15464c-28fc-41f7-806d-ed8470256cc3","added_by":"auto","created_at":"2025-10-27 13:35:04","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3305696,"visible":true,"origin":"","legend":"","description":"","filename":"fedexRevisedfinal.docx","url":"https://assets-eu.researchsquare.com/files/rs-7818007/v1/40e5f7d9773b065225cf3ed6.docx"},{"id":94376537,"identity":"e22b215e-acad-40b3-bc86-26fa0e80a258","added_by":"auto","created_at":"2025-10-27 13:35:49","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4714,"visible":true,"origin":"","legend":"","description":"","filename":"753046b89ac8427898c6bbd4f53926b5.json","url":"https://assets-eu.researchsquare.com/files/rs-7818007/v1/495b55c9b323d0f46d44cd3f.json"},{"id":94376347,"identity":"0dac691c-3f83-46ba-80d4-b8bde892ec79","added_by":"auto","created_at":"2025-10-27 13:35:30","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":266014,"visible":true,"origin":"","legend":"","description":"","filename":"suppletaryfedexfinal.docx","url":"https://assets-eu.researchsquare.com/files/rs-7818007/v1/076b3f545b21b6561ccd5e0e.docx"},{"id":94376532,"identity":"c3637ab1-b4cf-4e65-8bd2-5e570acc5dd5","added_by":"auto","created_at":"2025-10-27 13:35:49","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":133340,"visible":true,"origin":"","legend":"","description":"","filename":"753046b89ac8427898c6bbd4f53926b51enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7818007/v1/49a9a4510d92b9234728fa51.xml"},{"id":94376183,"identity":"823dd3aa-09b8-4c1a-b699-e16ac4921f0d","added_by":"auto","created_at":"2025-10-27 13:35:01","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1018549,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7818007/v1/ddc1451f4771b2b1a03d46d5.png"},{"id":94376535,"identity":"f6631884-1d12-4283-92a1-ffd5e42bc06f","added_by":"auto","created_at":"2025-10-27 13:35:49","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":775751,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7818007/v1/9b15d9503b113aa2acfa1da1.png"},{"id":94376817,"identity":"9390669d-0085-42f8-8873-593f0703222e","added_by":"auto","created_at":"2025-10-27 13:36:09","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":726452,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7818007/v1/6c8fe78f38bf7e0370fd1f3f.png"},{"id":94376491,"identity":"ae9eec60-0f89-4f9b-9747-c735efb2556d","added_by":"auto","created_at":"2025-10-27 13:35:44","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":706597,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7818007/v1/e1d4c0a2e76f15b380d7ece6.png"},{"id":94376732,"identity":"3dee3973-7ed0-4e25-845f-c51496651eb6","added_by":"auto","created_at":"2025-10-27 13:36:07","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":42222,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7818007/v1/5f831bb0038d284f620624fa.png"},{"id":94376354,"identity":"8d998b45-1e57-4f48-ae16-8ba1148fdd9b","added_by":"auto","created_at":"2025-10-27 13:35:30","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":37505,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7818007/v1/5050e1b0d842caeb80a34f77.png"},{"id":94376618,"identity":"f3516006-d5e3-4987-a028-26943a2d7703","added_by":"auto","created_at":"2025-10-27 13:35:55","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":42687,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7818007/v1/f2466ccae0cb03acad94240f.png"},{"id":94376345,"identity":"6d37adf3-84ae-4efb-8e8e-82342ddb541c","added_by":"auto","created_at":"2025-10-27 13:35:29","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":33709,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7818007/v1/f86f89ee67461ca977ed4365.png"},{"id":94376293,"identity":"421ba243-45d7-4d7d-8acb-9892ec315fe1","added_by":"auto","created_at":"2025-10-27 13:35:20","extension":"xml","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":133540,"visible":true,"origin":"","legend":"","description":"","filename":"753046b89ac8427898c6bbd4f53926b51structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7818007/v1/7d0d1e2bdf67767066f34a0e.xml"},{"id":94376196,"identity":"a3779a13-c6f2-4d6b-8039-6093d5da71c6","added_by":"auto","created_at":"2025-10-27 13:35:04","extension":"html","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":149394,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7818007/v1/82252c9b25feafcb7c281c4b.html"},{"id":94490609,"identity":"31fcd5ef-124b-4ffc-b068-a08307fc8570","added_by":"auto","created_at":"2025-10-27 17:12:58","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":658899,"visible":true,"origin":"","legend":"","description":"","filename":"fedexRevisedfinal.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7818007/v1_covered_7cb4a159-2a61-4b52-aac2-c5beb54e40b3.pdf"},{"id":94376385,"identity":"224f5e57-38c4-48d5-9445-85b5feda7dfd","added_by":"auto","created_at":"2025-10-27 13:35:34","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":266014,"visible":true,"origin":"","legend":"","description":"","filename":"suppletaryfedexfinal.docx","url":"https://assets-eu.researchsquare.com/files/rs-7818007/v1/744af0cfd652448d7ecede19.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"EcoFedX- Adaptive Multi-Objective Federated Learning for Energy- Efficient Image and Speech Signal Processing at the Edge ","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"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":"signal-image-and-video-processing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"sivp","sideBox":"Learn more about [Signal, Image and Video Processing](http://link.springer.com/journal/11760)","snPcode":"11760","submissionUrl":"https://submission.nature.com/new-submission/11760/3","title":"Signal, Image and Video Processing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Federated Learning, Multi-Objective Optimization, Edge AI, Signal Processing, Image and Speech Recognition, Energy Efficiency, TinyML, Adaptive Optimization","lastPublishedDoi":"10.21203/rs.3.rs-7818007/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7818007/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eFederated learning (FL) enables distributed model training across edge devices while preserving data privacy, yet existing methods struggle to balance accuracy, communication cost, and energy efficiency under heterogeneous conditions. This paper introduces EcoFedX, an adaptive multi-objective FL framework that jointly optimizes these three objectives through entropy- and energy-aware weighting of local updates.\u003cbr\u003e\nEcoFedX dynamically adjusts client contributions based on data diversity and device power profiles, achieving both statistical robustness and system-level sustainability.\u003cbr\u003e\nExtensive experiments on CIFAR-10, Google Speech Commands, and UCI-HAR datasets demonstrate that EcoFedX achieves ≈4 % higher accuracy, ≈38 ± 2 % lower communication cost, and ≈30 ± 3 % lower energy consumption relative to FedAvg, without compromising convergence.\u003cbr\u003e\nAblation and sensitivity analyses further validate the framework’s stability under varying compression ratios, client heterogeneity, and energy budgets (Supplementary S7). The results confirm that EcoFedX offers a principled, interpretable, and resource-efficient foundation for scalable edge intelligence.\u003c/p\u003e","manuscriptTitle":"EcoFedX- Adaptive Multi-Objective Federated Learning for Energy- Efficient Image and Speech Signal Processing at the Edge ","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-25 04:49:47","doi":"10.21203/rs.3.rs-7818007/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewersInvited","content":"","date":"2025-10-12T04:36:12+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-10T11:37:49+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-10-10T11:36:19+00:00","index":"","fulltext":""},{"type":"submitted","content":"Signal, Image and Video Processing","date":"2025-10-09T13:12:20+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"signal-image-and-video-processing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"sivp","sideBox":"Learn more about [Signal, Image and Video Processing](http://link.springer.com/journal/11760)","snPcode":"11760","submissionUrl":"https://submission.nature.com/new-submission/11760/3","title":"Signal, Image and Video Processing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"34d435fc-1a88-4d96-aadb-4deba180ee63","owner":[],"postedDate":"October 25th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2025-10-25T04:49:47+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-25 04:49:47","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7818007","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7818007","identity":"rs-7818007","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}