{"title":"利用深度神经网络和拥抱用户意图优化设计促进功率的太赫兹超大规模集成电路测试","authors":"T. Dharanika, J. Jaya, E. Nandakumar","doi":"10.1166/jno.2024.3619","DOIUrl":null,"url":null,"abstract":"VLSI (Very Large-Scale Integration) testing is a crucial step in ensuring the reliability and functionality of integrated circuits. However, conventional testing methods often lack the ability to address user-specific requirements, resulting in suboptimal outcomes. Terahertz technology\n offers unique capabilities for non-destructive testing, yet its integration with VLSI testing methodologies remains limited. Additionally, the neglect of user preferences in testing processes poses a challenge to tailoring testing procedures to specific user needs. This research presents a\n novel approach for fostered power terahertz VLSI testing, integrating deep neural networks (DNNs) and user intent optimization principles. The proposed framework comprises three main components: terahertz signal processing, deep neural network-based feature extraction, and user intent optimization.\n Terahertz signals are analyzed using deep neural networks trained on labeled datasets, while user intent optimization algorithms dynamically adjust testing parameters based on user feedback. Comparative analysis with traditional testing methods reveals superior testing coverage and accuracy\n achieved through the integration of Terahertz Technology (TZT), deep neural networks, and user intent optimization. Our approach demonstrated a significant improvement in reliability, with values ranging from 92.5% to 95.8%, depending on the specific testing scenario and dataset used for evaluation.\n The accuracy achieved by our methodology surpassed existing technologies by a substantial margin. Across various experiments, accuracy values ranged from 87.3% to 91.6%, indicating a consistent improvement over baseline methods.","PeriodicalId":16446,"journal":{"name":"Journal of Nanoelectronics and Optoelectronics","volume":null,"pages":null},"PeriodicalIF":0.6000,"publicationDate":"2024-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Design of Fostered Power Terahertz VLSI Testing Using Deep Neural Network and Embrace User Intent Optimization\",\"authors\":\"T. Dharanika, J. Jaya, E. Nandakumar\",\"doi\":\"10.1166/jno.2024.3619\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"VLSI (Very Large-Scale Integration) testing is a crucial step in ensuring the reliability and functionality of integrated circuits. However, conventional testing methods often lack the ability to address user-specific requirements, resulting in suboptimal outcomes. Terahertz technology\\n offers unique capabilities for non-destructive testing, yet its integration with VLSI testing methodologies remains limited. Additionally, the neglect of user preferences in testing processes poses a challenge to tailoring testing procedures to specific user needs. This research presents a\\n novel approach for fostered power terahertz VLSI testing, integrating deep neural networks (DNNs) and user intent optimization principles. The proposed framework comprises three main components: terahertz signal processing, deep neural network-based feature extraction, and user intent optimization.\\n Terahertz signals are analyzed using deep neural networks trained on labeled datasets, while user intent optimization algorithms dynamically adjust testing parameters based on user feedback. Comparative analysis with traditional testing methods reveals superior testing coverage and accuracy\\n achieved through the integration of Terahertz Technology (TZT), deep neural networks, and user intent optimization. Our approach demonstrated a significant improvement in reliability, with values ranging from 92.5% to 95.8%, depending on the specific testing scenario and dataset used for evaluation.\\n The accuracy achieved by our methodology surpassed existing technologies by a substantial margin. Across various experiments, accuracy values ranged from 87.3% to 91.6%, indicating a consistent improvement over baseline methods.\",\"PeriodicalId\":16446,\"journal\":{\"name\":\"Journal of Nanoelectronics and Optoelectronics\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.6000,\"publicationDate\":\"2024-07-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Nanoelectronics and Optoelectronics\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://doi.org/10.1166/jno.2024.3619\",\"RegionNum\":4,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"ENGINEERING, ELECTRICAL & ELECTRONIC\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Nanoelectronics and Optoelectronics","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1166/jno.2024.3619","RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}
Design of Fostered Power Terahertz VLSI Testing Using Deep Neural Network and Embrace User Intent Optimization
VLSI (Very Large-Scale Integration) testing is a crucial step in ensuring the reliability and functionality of integrated circuits. However, conventional testing methods often lack the ability to address user-specific requirements, resulting in suboptimal outcomes. Terahertz technology
offers unique capabilities for non-destructive testing, yet its integration with VLSI testing methodologies remains limited. Additionally, the neglect of user preferences in testing processes poses a challenge to tailoring testing procedures to specific user needs. This research presents a
novel approach for fostered power terahertz VLSI testing, integrating deep neural networks (DNNs) and user intent optimization principles. The proposed framework comprises three main components: terahertz signal processing, deep neural network-based feature extraction, and user intent optimization.
Terahertz signals are analyzed using deep neural networks trained on labeled datasets, while user intent optimization algorithms dynamically adjust testing parameters based on user feedback. Comparative analysis with traditional testing methods reveals superior testing coverage and accuracy
achieved through the integration of Terahertz Technology (TZT), deep neural networks, and user intent optimization. Our approach demonstrated a significant improvement in reliability, with values ranging from 92.5% to 95.8%, depending on the specific testing scenario and dataset used for evaluation.
The accuracy achieved by our methodology surpassed existing technologies by a substantial margin. Across various experiments, accuracy values ranged from 87.3% to 91.6%, indicating a consistent improvement over baseline methods.