CRII: CNS: Integrating Security Tasks into Multicore Real-Time Systems

该奖项全部或部分由2021年美国救援计划法案（公法117-2）资助。现代社会的许多关键系统（例如，汽车中的发动机控制单元、制造和发电厂中的逻辑控制器、飞行器控制和导航系统、工业控制系统、机器人辅助医疗保健中的传感和感知系统）具有“实时”（即，严格的时间和安全性）要求。新兴的物联网特定应用（例如，互联的自动驾驶汽车、无人驾驶飞行器）、现成组件的趋势使用以及广泛的互联网/网络连接性增加了这些关键系统中安全漏洞的可能性，正如最近的现实世界攻击所揭示的那样。这项研究的关键创新是开发一个统一的框架，将监控和检测机制作为一流的元素集成到实时系统的设计中，特别是那些使用多核芯片构建的系统。该项目将（a）设计新颖的算法、调度模型和框架，以将安全性集成到认识到实时要求的多核平台中，（B）构建设计时工具和系统级插件，以将所提出的技术集成到现成的系统中，以及（c）开发度量标准，以仔细权衡两个相互竞争的要求：及时性和安全性。这些想法将通过在两个现成平台上进行实验和测试来验证：多地形漫游车和六自由度机械臂。这项研究将通过使系统设计人员更好地了解如何整合安全问题来推进该领域，重点是揭示安全权衡，以确保对实时属性的干扰最小（或没有）。作为该项目的一部分开发的技术将使安全关键的实时系统更加安全，并适用于各种领域（例如，汽车、航空电子设备、无人机、太空漫游器、电网、制造工厂、医疗设备、工业控制系统）。拟议的研究和教育计划将提高下一代技术劳动力在网络物理系统和网络安全方面的知识。该奖项支持研究生和本科生的培训，威奇托州立大学新的安全课程的开发，以及研究成果融入教育材料。所有硬件、软件和系统实现（包括文档和教程）都将在公共存储库（https：//github.com/CPS2RL）中免费提供，供教育工作者、科学家、行业人员和业余爱好者访问和使用。该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

This award is funded in whole or in part under the American Rescue Plan Act of 2021 (Public Law 117-2).Many critical systems of modern society (e.g., engine control units in automobiles, logic controllers in manufacturing and power plants, aircraft control and navigation systems, industrial control systems, sensing and perception systems in robot-aided healthcare) have "real-time" (i.e., strict timing and safety) requirements. Emerging Internet-of-things-specific applications (e.g., connected autonomous cars, unmanned aerial vehicles), the trending use of off-the-shelf components, and widespread Internet/network connectivity expand the possibility of security breaches in those critical systems, as revealed by recent real-world attacks. The key innovation of this research is the development of a unified framework to integrate monitoring and detection mechanisms as first-class elements within the design of real-time systems, especially those built with multicore chips. This project will (a) devise novel algorithms, scheduling models, and frameworks to integrate security into multicore platforms that are cognizant of real-time requirements, (b) build design-time tools and system-level plugins to incorporate the proposed techniques into off-the-shelf systems, and (c) develop metrics to carefully trade-off two contending requirements: timeliness and security. The ideas will be validated through experimentation and testing on two off-the-shelf platforms: a multi-terrain rover and a six-degree-of-freedom robotic arm.This research will advance the field by enabling system designers to better understand how to integrate security concerns, with a focus on revealing security trade-offs to ensure minimal (or no) perturbations on real-time properties. Techniques developed as part of this project will make safety-critical, real-time systems more secure and applicable to various domains (e.g., automobiles, avionics, drones, space rovers, power grids, manufacturing plants, medical devices, industrial control systems). The proposed research and educational plans will enhance the knowledge of the next-generation technological workforce in cyber-physical systems and cyber-security. This award supports the training of graduate and undergraduate students, the development of a new security course at Wichita State University, and the integration of research findings into educational materials. All hardware, software, and system implementations (including documentation and tutorials) will be freely available in a public repository (https://github.com/CPS2RL) for educators, scientists, industry personnel, and hobbyists to access and use.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.