SaTC: STARSS: Design of Low-Cost Memory-Based Security Primitives and Techniques for High-Volume Products

确保电子计算设备的高度安全性和可靠性是一项重大挑战。核心问题包括安全可靠的身份识别、身份验证和底层硬件的完整性检查。基于硬件的安全原语，如物理不可克隆函数(PUF)，就它们保证可靠操作和抵抗物理攻击所需的成本而言，仍然是一项正在开发的工作。本项目研究适用于大批量生产的安全原语，其中硬件安全原语的设计、集成和测试的允许成本很低。本项目中开发的技术利用内存层次结构中的现有资源来同时涵盖多个硬件安全原语。本项目还通过在多个测试平台上的大量实验，验证了所提出的原语和技术的有效性、成本和攻击漏洞。为社会带来的好处包括为医疗保健提供更可靠、可靠和安全的电子验证，为情报、金融交易、能源和军事系统提供可靠的计算平台。

Ensuring a high level of security and reliability in the electronic computing devices is a significant challenge. Central issues include secure and reliable identification, authentication and integrity checking of underlying hardware. Hardware-based security primitives such as physical unclonable functions (PUFs) are still a work-in-progress in terms of the cost they require to guarantee reliable operation and their resistance to physical attacks. This project investigates security primitives suitable for high volume production where the allowable costs for design, integration, and testing of hardware security primitives are low. The techniques developed in this project utilize existing resources in the memory hierarchy to cover multiple hardware security primitives all at once. This project also verifies the effectiveness, cost, and attack vulnerabilities of the proposed primitives and techniques through extensive experiments on multiple test platforms. The benefits for society include more trustworthy, reliable, and secure verification of electronics for healthcare and dependable computing platforms for intelligence, financial transactions, energy, and military systems.