Parameter-Invariant Monitor Design for Cyber–Physical Systems

Abstract

The tight interaction between information technology and the physical world inherent in Cyber-Physical Systems (CPS) can challenge traditional approaches for monitoring safety and security. Data collected for robust CPS monitoring is often sparse and may lack rich training data describing critical events/attacks. Moreover, CPS often operate in diverse environments that can have significant inter/intra-system variability. Furthermore, CPS monitors that are not robust to data sparsity and inter/intra-system variability may result in inconsistent performance and may not be trusted for monitoring safety and security. Towards overcoming these challenges, this paper presents recent work on the design of parameter-invariant (PAIN) monitors for CPS. PAIN monitors are designed such that unknown events and system variability minimally affect the monitor performance. This work describes how PAIN designs can achieve a constant false alarm rate (CFAR) in the presence of data sparsity and intra/inter system variance in real-world CPS. To demonstrate the design of PAIN monitors for safety monitoring in CPS with different types of dynamics, we consider systems with networked dynamics, linear-time invariant dynamics, and hybrid dynamics that are discussed through case studies for building actuator fault detection, meal detection in type I diabetes, and detecting hypoxia caused by pulmonary shunts in infants. In all applications, the PAIN monitor is shown to have (significantly) less variance in monitoring performance and (often) outperforms other competing approaches in the literature. Finally, an initial application of PAIN monitoring for CPS security is presented along with challenges and research directions for future security monitoring deployments. Disciplines Computer Engineering | Computer Sciences Comments

Smart energy systems for sustainable smart cities: Current developments, trends and future directions

“…Though new opportunities have emerged with more pervasive sensing, historical data tends to be lacking in data describing critical events or attacks [89]. If a wide range of failure modes are possible, corresponding training data may be difficult or impossible to obtain in sufficient quantity.…”

“…A probabilistic neural network method was shown in [90] to out-perform more standard back-propagated neural networks for fault detection in refrigeration systems for example. Parameter-invariant approaches have been developed for classification in which constant false positive rates are obtained regardless of parameter values [89]. These approaches have been taken in networked systems [91] and for HVAC systems in the building energy domain [92].…”

Parameter Invariant Monitoring for Signal Temporal Logic

“…Authors in [26] have recently introduced a test procedure that can only be used for obtaining probabilistic beliefs about state of the system at single points in time (a.k.a. atomic propositions in STL), but is maximally invariant to nuisance parameters (e.g.…”

“…This intuitively means, not only this test procedure is able to ignore parts of information that are corrupted by the nuisance parameters, it only ignores a minimum amount of them. For example, suppose o 1 and o 2 are two observations that could have been affected by the nuisance parameters, and let o ′ 1 and o ′ 2 be what the algorithm in [26] obtains after filtering some of the information out. There are two more challenges in monitoring continuous time system behaviors against STL properties.…”

“…Our main contribution in this paper is to show how one can extend any parameter invariant test procedure for atomic propositions (including the one that is introduced in [26]) to a parameter invariant test procedure for efficiently monitoring STL properties over continuous time signals. Furthermore, if the test procedure for atomic propositions is maximally invariant then the test procedure that we construct will be maximally invariant as well.…”