Collaborative Research: CT-T: Manufacturing Variability-based Hardware Protection Techniques

合作研究：CT-T：基于制造变异性的硬件保护技术

• 批准号: 0716443
• 负责人: John Lach
• 金额: $ 5万
• 依托单位: University of Virginia Main Campus
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2007
• 资助国家: 美国
• 起止时间: 2007-10-01 至 2008-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0716443&HistoricalAwards=false
• 关键词: Collaborative; Research; CT; Manufacturing; Variability

It is estimated that hardware (HW) intellectual property piracy induces almost an order of magnitude higher loss when compared to software piracy cost and greatly facilitates both software and entertainment data piracy. In addition, HW vulnerabilities can be exploited for security attacks, especially the increasing threat of HW malware attacks (malicious or unintentional alterations of design specifications that compromise the correctness of the functionality under specific conditions) due to business models moving to external foundries and assembly houses. The strategic objective of this project is to give impetus to research on the protection against HW malware attacks and piracy with a multi-pronged attack that includes the development of both the conceptual foundation and several practical HW protection techniques.The intellectual merit of this project is the development of HW protection techniques that leverage the manufacturing variability (MV) inherent in modern fabrication processes and the resulting uniqueness of every manufactured integrated circuit (IC). This project is developing: (i) HW malware detection techniques considering only a single IC under very mild statistical assumptions about MV, and (ii) active HW metering protocols that prevent foundries from distributing unauthorized design copies by integrating MV into the functionality of the targeted design in such a way that only the designer can issue a key that unlocks the design after power up so that it becomes functional.The broader impacts of this project are that it will form the basis for a new industry that ensures HW integrity in a complex horizontal IC business model. In addition, this project is exposing both undergraduate and graduate students to these increasingly important topics and providing research opportunities for underrepresented students.

据估计，硬件（HW）知识产权盗版造成的损失几乎比软件盗版成本高一个数量级，极大地促进了软件和娱乐数据的盗版。此外，由于业务模型转移到外部铸造厂和装配厂，硬件漏洞可以被用于安全攻击，特别是硬件恶意软件攻击（恶意或无意地改变设计规范，损害特定条件下功能的正确性）的威胁日益增加。该项目的战略目标是通过多管齐下的攻击，推动对硬件恶意软件攻击和盗版的保护研究，包括概念基础和几种实用的硬件保护技术的发展。该项目的智力优势在于开发了利用现代制造工艺中固有的制造变异性（MV）和每个制造集成电路（IC）的独特性的HW保护技术。该项目正在开发：(i)硬件恶意软件检测技术，仅考虑在非常温和的MV统计假设下的单个IC，以及（ii）主动硬件计量协议，通过将MV集成到目标设计的功能中，防止代工厂分发未经授权的设计副本，这样只有设计师才能发出钥匙，在通电后解锁设计，使其具有功能。该项目的更广泛影响是，它将形成一个新产业的基础，确保硬件在复杂的水平集成电路商业模式中的完整性。此外，该项目使本科生和研究生都接触到这些日益重要的主题，并为代表性不足的学生提供研究机会。