SHF: Small: Fault Model Evaluation and Discovery

SHF：小型：故障模型评估和发现

• 批准号: 1816512
• 负责人: Ronald Blanton
• 金额: $ 44.63万
• 依托单位: Carnegie-Mellon University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-06-01 至 2024-05-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1816512&HistoricalAwards=false
• 关键词: SHF; Small; Fault; Model; Evaluation

Integrated electronic circuits (ICs) today are ubiquitous in that they affect every aspect of our daily lives, ranging from use within all forms of mobile devices (smart phones, laptops, etc.) to all modes of transportation (cars, aircraft, etc.). Advances in electronics also form the foundation for the recent and forthcoming advances in machine learning, artificial intelligence, autonomous vehicles, etc. The research work stemming from the project will aid the continued advancement of the electronic semiconductor industry by helping leading companies ensure that electronics work properly and reliably over long time periods. Additionally, the research conducted will include undergraduate student researchers, and the results will be disseminated to the research community, and incorporated into post-secondary curriculum.The research will specifically examine how electronic failures are modeled, and for any shortcomings discovered, new models that are more accurate will be developed. While it is widely known that model effectiveness greatly depends on the IC and its fabrication, it is not known which models will be proven ineffective beforehand. In addition, especially for cutting-edge technologies, it is expected that there also will be a need for new, more accurate fault models for state-of-the-art ICs. Therefore, the objective of this research is to develop a comprehensive data-driven methodology for the continuous evaluation and development of fault models. Data from failed ICs will be analyzed to understand the likelihood and types of failures. For the expected mismatch between existing models and measured data, new models will be proposed and evaluated. Successfully meeting the objective of this research would will enable IC manufacturers to produce more reliable ICs at much lower cost.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

今天的集成电子电路（IC）无处不在，因为它们影响我们日常生活的各个方面，从在所有形式的移动的设备（智能电话、膝上型计算机等）内的使用，适用于所有运输方式（汽车、飞机等）。电子技术的进步也为机器学习、人工智能、自动驾驶汽车等领域最近和即将取得的进展奠定了基础，该项目的研究工作将通过帮助领先公司确保电子设备长期正常可靠地工作，帮助电子半导体行业的持续发展。此外，研究人员还将包括本科生，研究结果将被传播到研究界，并纳入高等教育课程。研究将专门研究电子故障的建模方法，并针对发现的任何缺陷，开发更准确的新模型。虽然众所周知，模型的有效性在很大程度上取决于IC及其制造，但不知道哪些模型将被证明是无效的。此外，特别是对于尖端技术，预计还将需要用于最先进IC的新的、更准确的故障模型。因此，本研究的目标是开发一个全面的数据驱动的方法，用于故障模型的持续评估和开发。将分析故障IC的数据，以了解故障的可能性和类型。对于现有模型和测量数据之间的预期不匹配，将提出新模型并进行评估。成功实现这项研究的目标将使集成电路制造商能够以更低的成本生产更可靠的集成电路。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

IPSA: Integer Programming via Sparse Approximation for Efficient Test-Chip Design

IPSA：通过稀疏近似进行整数编程以实现高效的测试芯片设计

• DOI: 10.1109/iccd46524.2019.00011
• 发表时间: 2019
• 期刊: 2019 IEEE 37th International Conference on Computer Design (ICCD
• 作者: Huang, Qicheng;Fang, Chenlei;Liu, Zeye;Ding, Ruizhou;Blanton, R. D.
• 通讯作者: Blanton, R. D.

A Deterministic-Statistical Multiple-Defect Diagnosis Methodology

• DOI: 10.1109/vts48691.2020.9107603
• 发表时间: 2020-04
• 期刊: 2020 IEEE 38th VLSI Test Symposium (VTS)
• 作者: Soumya Mittal;R. D. Blanton

Improving Test Chip Design Efficiency via Machine Learning

通过机器学习提高测试芯片设计效率

• DOI: 10.1109/itc44170.2019.9000131
• 发表时间: 2019
• 期刊: 2019 IEEE International Test Conference
• 作者: Liu, Zeye;Huang, Qicheng;Fang, Chenlei;Blanton, R. D.
• 通讯作者: Blanton, R. D.

LearnX: A Hybrid Deterministic-Statistical Defect Diagnosis Methodology

LearnX：混合确定性统计缺陷诊断方法

• 发表时间: 2019
• 期刊: Proceedings
• 作者: Mittal, Soumya;Blanton, Ronald
• 通讯作者: Blanton, Ronald

Adaptive Test Pattern Reordering for Diagnosis using k-Nearest Neighbors

使用 k 最近邻进行自适应测试模式重新排序以进行诊断

• DOI: 10.1109/itc-asia51099.2020.00022
• 发表时间: 2020
• 期刊: 2020 IEEE International Test Conference in Asia (ITC-Asia
• 作者: Fang, Chenlei;Huang, Qicheng;Blanton, R.D.
• 通讯作者: Blanton, R.D.