Unsupervised Deep Learning for Binary Offloading in Mobile Edge Computation Network

Unsupervised Deep Learning for Binary Offloading in Mobile Edge Computation Network | 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 Unsupervised Deep Learning for Binary Offloading in Mobile Edge Computation Network Xue Chen, Hongbo Xu, Guoping Zhang, Yun Chen, Ruijie Li This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-372831/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 02 Dec, 2021 Read the published version in Wireless Personal Communications → Version 1 posted 4 You are reading this latest preprint version Abstract Mobile edge computation (MEC) is a potential technology to reduce the energy consumption and task execution delay for tackling computation-intensive tasks on mobile device (MD). The resource allocation of MEC is an optimization problem, however, the existing large amount of computation may hinder its practical application. In this work, we propose a multiuser MEC framework based on unsupervised deep learning (DL) to reduce energy consumption and computation by offloading tasks to edge servers. The binary offloading decision and resource allocation are jointly optimized to minimize energy consumption of MDs under latency constraint and transmit power constraint. This joint optimization problem is a mixed integer nonconvex problem which result in the gradient vanishing problem in backpropagation. To address this, we propose a novel binary computation offloading scheme (BCOS), in which a deep neural network (DNN) with an auxiliary network is designed. By using the auxiliary network as a teacher network, the student network can obtain the lossless gradient information in joint training phase. As a result, the sub-optimal solution of the optimization problem can be acquired by the learning-based BCOS. Simulation results demonstrate that the BCOS is effective to solve the binary offloading problem by the trained network with low complexity. Scientific Communication Technical Communication Electrical Engineering MEC Binary offloading decision Unsupervised deep learning Auxiliary network Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Full Text Due to technical limitations, full-text HTML conversion of this manuscript could not be completed. However, the manuscript can be downloaded and accessed as a PDF. Cite Share Download PDF Status: Published Journal Publication published 02 Dec, 2021 Read the published version in Wireless Personal Communications → Version 1 posted Reviewers invited by journal 10 Apr, 2021 Reviews received at journal 10 Apr, 2021 Editor assigned by journal 27 Mar, 2021 First submitted to journal 24 Mar, 2021 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-372831","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":20928752,"identity":"79521964-c78c-47a6-878c-77d8c86e3b12","order_by":0,"name":"Xue Chen","email":"","orcid":"","institution":"Central China Normal University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Xue","middleName":"","lastName":"Chen","suffix":""},{"id":20928753,"identity":"c67c3c14-48cd-4df0-947f-cbac7a6e6b66","order_by":1,"name":"Hongbo Xu","email":"","orcid":"","institution":"Central China Normal University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Hongbo","middleName":"","lastName":"Xu","suffix":""},{"id":20928754,"identity":"34ae36aa-2f0a-482f-8f0c-c3189ce38f38","order_by":2,"name":"Guoping Zhang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+UlEQVRIiWNgGAWjYHACNhAhx8DA2HgAJiRBjBZjoJYG0rQkNgAJ4rTIR7c/e/BzR2362vbDQFtq7PIMDjAfvM3DYJeHS4vhnQPphr1njuduO5MI1HIsudjgAFuyNQ9DcjFOLTMSjknwth3L3XYAqAXoncQNB3jMpHkYDoCdil1LYpvk37Zj6WbnH8K08H/Dq0VeIplNmretJsHsBsIWNrxaDGSOsUnLth0w3HYDaEvCseTEmYfZjC3nGCTjtmV2+zPJt2118mbn0x8++FBjl9h3vPnhjTcVdrhtuQGmDkN4CSCCGSyOQz3Ilhlgqg63ilEwCkbBKBgFABmyYKne37AUAAAAAElFTkSuQmCC","orcid":"","institution":"Central China Normal University","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Guoping","middleName":"","lastName":"Zhang","suffix":""},{"id":20928755,"identity":"6508a16a-8875-4b5f-be51-6f69efec4217","order_by":3,"name":"Yun Chen","email":"","orcid":"","institution":"Central China Normal University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yun","middleName":"","lastName":"Chen","suffix":""},{"id":20928756,"identity":"0cb66925-18e8-4fed-ba1c-5198e4692d23","order_by":4,"name":"Ruijie Li","email":"","orcid":"","institution":"Central China Normal University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Ruijie","middleName":"","lastName":"Li","suffix":""}],"badges":[],"createdAt":"2021-03-29 05:26:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-372831/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-372831/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s11277-021-09433-9","type":"published","date":"2021-12-02T09:49:29+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":8000787,"identity":"d3ba369b-92d0-4119-99fb-e519af5f1dc3","added_by":"auto","created_at":"2021-04-14 12:20:49","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":44798,"visible":true,"origin":"","legend":"Multiuser MECO system.","description":"","filename":"fig1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-372831/v1/a318d1e90f6741153a0e9dad.jpg"},{"id":8000599,"identity":"8c9ddc06-0ab8-4382-9094-e8ef0ab6ea91","added_by":"auto","created_at":"2021-04-14 12:17:49","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":88190,"visible":true,"origin":"","legend":"A fully connected network with 2 hidden layers.","description":"","filename":"fig2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-372831/v1/654d5f645e91c1f4c41613e7.jpg"},{"id":8000784,"identity":"06e4159e-efca-482d-b8b2-cfe2c0819ea0","added_by":"auto","created_at":"2021-04-14 12:20:49","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":67549,"visible":true,"origin":"","legend":"Schematic of the proposed joint training with auxiliary DNN. The FCN of offloading\ndecision is shared by the auxiliary teacher network and the student network. In training\nphase, the parameters of the auxiliary teacher network and the student network are updated alternatively. Then, the shared FCN can obtain the lossless gradient information during backpropagation.","description":"","filename":"fig3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-372831/v1/0588aa7b1e417fc6db46958b.jpg"},{"id":8000602,"identity":"79d2fc61-6c1e-4df6-9f21-e583540a97d4","added_by":"auto","created_at":"2021-04-14 12:17:49","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":54006,"visible":true,"origin":"","legend":"The comparison of weighted sum mobile energy consumption between MCOS, PCOS\nand BCOS with the increasement of mobile number from 4 to 20, where T = 0:5s.","description":"","filename":"fig4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-372831/v1/a19b7fd3b447622dcdb84ea5.jpg"},{"id":8000786,"identity":"9aba7057-e0b7-4231-9798-35662dfeb37e","added_by":"auto","created_at":"2021-04-14 12:20:49","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":51029,"visible":true,"origin":"","legend":"The comparison of weighted sum mobile energy consumption between MCOS, PCOS\nand BCOS with the increasement of time slot T from 0.3s to 1.6s, where the mobile number is 14.","description":"","filename":"fig5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-372831/v1/ebd83eabd2a2cc5e35d730c6.jpg"},{"id":8000833,"identity":"d9ca6c51-0ff4-471d-9ba6-cde0358128f6","added_by":"auto","created_at":"2021-04-14 12:23:49","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":54142,"visible":true,"origin":"","legend":"CDF of the sum of realistic task offloading time for K MDs where T = 1s and K = 14.","description":"","filename":"fig6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-372831/v1/2e462eef935cc391d7a8aa6c.jpg"},{"id":16102210,"identity":"e10f662b-bbd9-4db6-8447-a7d1802e32ba","added_by":"auto","created_at":"2021-12-02 09:49:36","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2784240,"visible":true,"origin":"","legend":"","description":"","filename":"template.pdf","url":"https://assets-eu.researchsquare.com/files/rs-372831/v1_covered.pdf"},{"id":13622013,"identity":"aea34dff-efb3-4ace-9c8b-3939aeee3e11","added_by":"auto","created_at":"2021-09-17 07:12:59","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2778963,"visible":true,"origin":"","legend":"","description":"","filename":"template.pdf","url":"https://assets-eu.researchsquare.com/files/rs-372831/v1_covered.pdf"},{"id":8000880,"identity":"d66592f8-0408-4b3c-8964-a8c7bdf6219c","added_by":"auto","created_at":"2021-04-14 12:26:56","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2577308,"visible":true,"origin":"","legend":"","description":"","filename":"template.pdf","url":"https://assets-eu.researchsquare.com/files/rs-372831/v1_stamped.pdf"}],"financialInterests":"","formattedTitle":"Unsupervised Deep Learning for Binary Offloading in Mobile Edge Computation Network","fulltext":[{"header":"Full Text","content":"\u003cp\u003eDue to technical limitations, full-text HTML conversion of this manuscript could not be completed. However, the manuscript can be downloaded and accessed as a PDF.\u003c/p\u003e"}],"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":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"wireless-personal-communications","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"wire","sideBox":"Learn more about [Wireless Personal Communications](https://www.springer.com/journal/11277)","snPcode":"11277","submissionUrl":"https://submission.nature.com/new-submission/11277/3","title":"Wireless Personal Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"MEC, Binary offloading decision, Unsupervised deep learning, Auxiliary network","lastPublishedDoi":"10.21203/rs.3.rs-372831/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-372831/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Mobile edge computation (MEC) is a potential technology to reduce the energy consumption and task execution delay for tackling computation-intensive tasks on mobile device (MD). The resource allocation of MEC is an optimization problem, however, the existing large amount of computation may hinder its practical application. In this work, we propose a multiuser MEC framework based on unsupervised deep learning (DL) to reduce energy consumption and computation by offloading tasks to edge servers. The binary offloading decision and resource allocation are jointly optimized to minimize energy consumption of MDs under latency constraint and transmit power constraint. This joint optimization problem is a mixed integer nonconvex problem which result in the gradient vanishing problem in backpropagation. To address this, we propose a novel binary computation offloading scheme (BCOS), in which a deep neural network (DNN) with an auxiliary network is designed. By using the auxiliary network as a teacher network, the student network can obtain the lossless gradient information in joint training phase. As a result, the sub-optimal solution of the optimization problem can be acquired by the learning-based BCOS. Simulation results demonstrate that the BCOS is effective to solve the binary offloading problem by the trained network with low complexity.","manuscriptTitle":"Unsupervised Deep Learning for Binary Offloading in Mobile Edge Computation Network","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2021-04-14 12:17:47","doi":"10.21203/rs.3.rs-372831/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewersInvited","content":"","date":"2021-04-11T00:00:00+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2021-04-11T00:00:00+00:00","index":0,"fulltext":""},{"type":"editorAssigned","content":"","date":"2021-03-28T00:00:00+00:00","index":"","fulltext":""},{"type":"submitted","content":"Wireless Personal Communications","date":"2021-03-24T07:36:41+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"wireless-personal-communications","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"wire","sideBox":"Learn more about [Wireless Personal Communications](https://www.springer.com/journal/11277)","snPcode":"11277","submissionUrl":"https://submission.nature.com/new-submission/11277/3","title":"Wireless Personal Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"71a20b44-c484-4eaa-8e4c-0dee60a06c58","owner":[],"postedDate":"April 14th, 2021","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":3616405,"name":"Scientific Communication"},{"id":3616406,"name":"Technical Communication"},{"id":3616407,"name":"Electrical Engineering"}],"tags":[],"updatedAt":"2021-12-02T09:49:29+00:00","versionOfRecord":{"articleIdentity":"rs-372831","link":"https://doi.org/10.1007/s11277-021-09433-9","journal":{"identity":"wireless-personal-communications","isVorOnly":false,"title":"Wireless Personal Communications"},"publishedOn":"2021-12-02 09:49:29","publishedOnDateReadable":"December 2nd, 2021"},"versionCreatedAt":"2021-04-14 12:17:47","video":"","vorDoi":"10.1007/s11277-021-09433-9","vorDoiUrl":"https://doi.org/10.1007/s11277-021-09433-9","workflowStages":[]},"version":"v1","identity":"rs-372831","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-372831","identity":"rs-372831","version":["v1"]},"buildId":"cBFmMYwuxLRRLfASyISRj","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}