CAREER: Practical Control Engineering Principles to Improve the Security and Privacy of Cyber-Physical Systems

该项目通过整合信息安全社区的理论和最佳实践以及控制理论社区的实践方法，重点解决网络物理系统（CPS）的安全和隐私问题。该项目的第一部分侧重于网络物理关键基础设施（如电网、配水网络和运输网络）的安全和保护，以防止可能导致服务中断、基础设施损坏甚至生命损失的计算机攻击。该项目的第二部分侧重于CPS的隐私，并提出了新的算法来处理人类活动中前所未有的数据收集粒度。这两部分的工作重点是将实际控制理论概念集成到计算机安全解决方案中。特别是，在过去十年中，控制理论界提出了CPS安全的基本进展；与此同时，电脑保安界也在CPS保安及私隐的实际实施方面取得重大进展。虽然这两个领域各自独立地取得了重大进展，但两个领域之间仍然存在很大的语言和概念障碍，因此，计算机安全专家已经从控制理论研究人员那里制定了一个平行和独立的研究议程。为了设计未来的CPS安全和隐私机制，这两个社区需要更紧密地联系在一起，并利用各自开发的见解。本项目试图通过在两个研究问题中利用受控系统的物理特性来促进这两个社区的整合：(1)基于物理的CPS安全；(2)基于物理的CPS隐私。基于物理的CPS安全利用传感器和控制信号的时间序列来检测与预期操作的偏差。这是安全和控制理论场地的一个不断发展的研究领域，尽管在这个领域有几个开放的问题。本建议解决了其中一些悬而未决的问题，包括定义新的评估指标，以捕获控制系统的独特操作属性，对模型和异常检测测试的不同建议进行一致的评估，以及开发新的工业控制协议解析器。基于物理的CPS隐私侧重于如何将一般隐私建议（如公平信息实践原则）引导到网络物理系统中，利用这些物理系统通常有一个目标要实现的事实，而这个目标取决于运营商的数据处理政策。该项目侧重于调查隐私和控制性能之间的权衡，并开发工具来指导如何实现数据最小化、数据延迟和数据保留。

This project focuses on tackling the security and privacy of Cyber-Physical Systems (CPS) by integrating the theory and best practices from the information security community as well as practical approaches from the control theory community. The first part of the project focuses on security and protection of cyber-physical critical infrastructures such as the power grid, water distribution networks, and transportation networks against computer attacks in order to prevent disruptions that may cause loss of service, infrastructure damage or even loss of life. The second part of the project focuses on privacy of CPS and proposes new algorithms to deal with the unprecedented levels of data collection granularity of physical human activity. The work in these two parts focuses on the integration of practical control theory concepts into computer security solutions. In particular, in the last decade, the control theory community has proposed fundamental advances in CPS security; in parallel, the computer security community has also achieved significant advances in practical implementation aspects for CPS security and privacy. While both of these fields have made significant progress independently, there is still a large language and conceptual barrier between the two fields, and as a result, computer security experts have developed a parallel and independent research agenda from control theory researchers. In order to design future CPS security and privacy mechanisms, the two communities need to come closer together and leverage the insights that each has developed. This project attempts to facilitate the integration of these two communities by leveraging the physical properties of the system under control in two research problems: (1) Physics-based CPS security; and (2) Physics-based CPS privacy.Physics-based CPS security leverages the time series from sensor and control signals to detect deviations from expected operation. This is a growing area of research in both security and control theory venues, although there are several open problems in this space. This proposal tackles some of these open problems including the definition of new evaluation metrics that capture the unique operational properties of control systems, the consistent evaluation of different proposals for models and anomaly detection tests, and the development of new industrial control protocol parsers. Physics-based CPS privacy focuses on how to guide the implementation of general privacy recommendations like the Fair Information Practice principles into cyber-physical systems, leveraging the fact that these physical systems often have an objective to achieve, and this objective depends on the data-handling policies of the operator. The project focuses on investigating the trade-off between privacy and control performance and developing tools to guide how data minimization, data delays, and data retention should be implemented.