Automatic Feature Selection Based Machine Learning Models for Hardware Trojan Detection

Automatic Feature Selection Based Machine Learning Models for Hardware Trojan Detection | 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 Automatic Feature Selection Based Machine Learning Models for Hardware Trojan Detection Shivam Dubey, Vijaypal Singh Rathor This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5018830/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 Hardware Trojan (HT) is a significant threat to the integrity and security of the Integrated Circuit (IC). Hardware Trojans can be implanted in any stage of IC development. Detection of HT is difficult during testing due to its stealthy nature and IC’s complex design. Therefore, researchers have focused on developing HT detection methods to mitigate the threats of HT. In recent years, machine learning (ML) has emerged as a promising approach that uses various Trojan-related circuit features for detecting the hardware Trojans. The previously proposed ML-based methods either use all known features or use manual/random feature selection. Due to this, these methods are complex and provide low performance. This work proposes a new ML-based method that encompasses different automatic feature selection algorithms, such as Principal Component Analysis (PCA) and AutoEncoder, for effective Trojan detection in ICs. We also use algorithms like Lasso, Ridge, and Elastic-net to select the best features based on their coefficient values. Using feature selection enables us to identify the most influential features for accurately detecting hardware Trojans. Further, we develop different ML models by integrating these automatic feature selection algorithms to achieve high performance (in terms of accuracy and F1-score) during HT detection. The experimental evaluation shows that the proposed Autoencoder achieves 99.5% average accuracy and 99.28% F1 score, and PCA provides on average 98.6% accuracy and 98.35% F1-Score with different proposed machine learning models. The proposed XGBoost, Random forest-based ML model achieves 100% accuracy and F1-score with Autoencoder while detecting HT in different datasets prepared using Trust-hub benchmarks. The proposed model exhibits a notable improvement in the F1-score compared to the best-known existing ML-based model, ranging from 0.1% to 2.1%. The proposed method’s top-performing model (RF with Autoencoder) achieves significant enhancements in the F1-score, ranging from 2% to 53%, over the performance of existing ML-based hub trust detection methods. Machine Learning Hardware Trojan Detection Feature Selection Dimensionality Reduction Regularization 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-5018830","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":354073169,"identity":"f1d969a9-2a21-442e-a13a-b77995728b97","order_by":0,"name":"Shivam Dubey","email":"","orcid":"","institution":"ABV-Indian Institute of Information Technology and Management","correspondingAuthor":false,"prefix":"","firstName":"Shivam","middleName":"","lastName":"Dubey","suffix":""},{"id":354073170,"identity":"40c7eb0d-ead8-4c51-b3ac-11e9e8bb8067","order_by":1,"name":"Vijaypal Singh Rathor","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAzklEQVRIiWNgGAWjYLACxgYGBn5mxoYDCC4hcBCoRrIZqOUASVoMQMoPEOMmg/NnH37+uMMm3/g4c+PhjzkM8vwNzG0P8Gq5kW4scfBMmuW2w0CHHdzGYDjjAGO7AT4tkjPYGCQOth02MINqYdzAwNgmgVdL/zHmHyAtxs0QLfYEtfAzpLGBbTFghmhJJKxFIo3N4mxbmoEEyGFnt0kkzzhMQAsb/zHmG5VtNgb8/ccff6jcZmPb397+DK8WdABUzEyK+lEwCkbBKBgFWAEAwVNMN0I77+4AAAAASUVORK5CYII=","orcid":"","institution":"ABV-Indian Institute of Information Technology and Management","correspondingAuthor":true,"prefix":"","firstName":"Vijaypal","middleName":"Singh","lastName":"Rathor","suffix":""}],"badges":[],"createdAt":"2024-09-02 13:43:57","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5018830/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5018830/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":66846914,"identity":"799f059b-91ab-4ff1-8f04-1f5e3376fe1e","added_by":"auto","created_at":"2024-10-17 06:09:03","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":418314,"visible":true,"origin":"","legend":"","description":"","filename":"barejrnlnewsample4.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5018830/v1_covered_1581f7a1-442d-40dc-8d0c-df8769e145d1.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Automatic Feature Selection Based Machine Learning Models for Hardware Trojan Detection","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":"Machine Learning, Hardware Trojan Detection, Feature Selection, Dimensionality Reduction, Regularization","lastPublishedDoi":"10.21203/rs.3.rs-5018830/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5018830/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eHardware Trojan (HT) is a significant threat to the integrity and security of the Integrated Circuit (IC). Hardware Trojans can be implanted in any stage of IC development. Detection of HT is difficult during testing due to its stealthy nature and IC’s complex design. Therefore, researchers have focused on developing HT detection methods to mitigate the threats of HT. In recent years, machine learning (ML) has emerged as a promising approach that uses various Trojan-related circuit features for detecting the hardware Trojans. The previously proposed ML-based methods either use all known features or use manual/random feature selection. Due to this, these methods are complex and provide low performance. This work proposes a new ML-based method that encompasses different automatic feature selection algorithms, such as Principal Component Analysis (PCA) and AutoEncoder, for effective Trojan detection in ICs. We also use algorithms like Lasso, Ridge, and Elastic-net to select the best features based on their coefficient values. Using feature selection enables us to identify the most influential features for accurately detecting hardware Trojans. Further, we develop different ML models by integrating these automatic feature selection algorithms to achieve high performance (in terms of accuracy and F1-score) during HT detection. The experimental evaluation shows that the proposed Autoencoder achieves 99.5% average accuracy and 99.28% F1 score, and PCA provides on average 98.6% accuracy and 98.35% F1-Score with different proposed machine learning models. The proposed XGBoost, Random forest-based ML model achieves 100% accuracy and F1-score with Autoencoder while detecting HT in different datasets prepared using Trust-hub benchmarks. The proposed model exhibits a notable improvement in the F1-score compared to the best-known existing ML-based model, ranging from 0.1% to 2.1%. The proposed method’s top-performing model (RF with Autoencoder) achieves significant enhancements in the F1-score, ranging from 2% to 53%, over the performance of existing ML-based hub trust detection methods.\u003c/p\u003e","manuscriptTitle":"Automatic Feature Selection Based Machine Learning Models for Hardware Trojan Detection","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-10-08 16:32:28","doi":"10.21203/rs.3.rs-5018830/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":"95f0bad7-9097-4dfb-85d0-a1c8a127b2d8","owner":[],"postedDate":"October 8th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-10-17T06:08:44+00:00","versionOfRecord":[],"versionCreatedAt":"2024-10-08 16:32:28","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5018830","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5018830","identity":"rs-5018830","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}