Finite‐time l2−l∞ filtering for persistent dwell‐time switched piecewise‐affine systems against deception attacks

• Published in: Applied mathematics and computation Vol. 427
• Main Authors: Mei, Zhen, Fang, Ting, Shen, Hao
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 15.08.2022
• Subjects: Deception attacks; Finite-time filtering; Persistent dwell-time; Piecewise-affine systems; Piecewise-affine systems; Finite-time filtering; Persistent dwell-time; Deception attacks
• ISSN: 0096-3003; 1873-5649
• DOI: 10.1016/j.amc.2022.127088

•Based on the mode and region information, a mode-dependent and region-dependent filter is constructed in this paper, which is more general than the mode-dependent filter.•Compared with the DT and ADT switching regularity, the prescribed PDT switching regularity, which has a stronger generality, is introduced to obtain the desired filter of switched PWA systems.•For a class of discrete-time PDT-switched PWA systems against DAs, as the first attempt, we investigate the issue of finite-time l2−l∞ filtering and benefits of the available filter design method are verified by two simulation examples in this paper. In this paper, the finite-time l2−l∞ filtering problem is investigated for discrete-time switched piecewise-affine systems against deception attacks, where switching signals are subject to the persistent dwell-time strategy. Considering the effect of deception attacks, a set of random variables obeying the Bernoulli distribution is used to describe the probability of false data injection between the plant and the filter. Based on the partition of state space with different operating regions, the objective of this paper is to design a mode-dependent and region-dependent filter such that the filtering error system is finite-time bounded and satisfies an l2−l∞ performance. By resorting to Lyapunov stability theory and finite-time analysis theory, some sufficient conditions are proposed to obtain such a desired filter. Finally, the availability of the proposed approach is demonstrated by two simulation examples.