Security-Based Distributionally Robust Optimization Self-Triggered SMPC for Constrained Cyber-Physical Systems Subject to Unknown Disturbances and Denial-of-Service Attacks

• Published in: IEEE internet of things journal Vol. 12; no. 12; pp. 21757 - 21769
• Main Authors: Chen, Yadong, Cheng, Peng
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 15.06.2025; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Chance constraint; Constraints; Control systems; Cyber-physical systems; Cybersecurity; Denial of service attacks; Discrete systems; distributionally robust optimization; Disturbances; Event detection; Internet of Things; Optimization; Prediction algorithms; Predictive control; Predictive models; Quadratic forms; Robustness (mathematics); Sampling; self-triggered stochastic model predictive control (MPC); Sensors; Stochastic models; Stochastic processes; Vectors
• ISSN: 2327-4662; 2327-4662
• DOI: 10.1109/JIOT.2025.3548405

In this article, we propose a distributionally robust optimization-based self-triggered stochastic model predictive control (DRSMPC) algorithm for linear discrete systems that are subject to unbounded stochastic disturbances and Denial-of-Service (DoS) attacks. Assuming that only the first and second moments of the disturbances are available, we transform the objective function into a compact quadratic form and reformulate the state chance constraint into second-order cone constraint, which enhance tractability during the solution process. To reduce communication and sampling times within the system, we introduce a self-triggering update framework, which computes the sampling instants and the control input sequences between consecutive sampling instants based on the sampled state. The main contribution of DRSMPC is to deal with unknown disturbances, reduce the unnecessary sampling also be able to defend against DoS attacks. Furthermore, we demonstrate that DRSMPC is recursively feasible and cyber-physical systems (CPSs) remaining quadratic stability. Numerical simulations validate the effectiveness of the proposed algorithm.