Event-based asynchronous and resilient filtering for Markov jump singularly perturbed systems against deception attacks

• Published in: ISA transactions 2021-06, Vol.112, p.56-73
• Main Authors: Wang, Yanqian, Chen, Fu, Zhuang, Guangming, Song, Gongfei
• Format: Article
• Language: English
• Subjects: Cyber–physical systems; Deception-attacks; Dynamic event-based transmission protocol; Linear matrix inequality; Markov jump singularly perturbed systems
• ISSN: 0019-0578; 1879-2022
• DOI: 10.1016/j.isatra.2020.11.029

The dynamic event-based asynchronous and resilient dissipative filter design for Markov jump singularly perturbed systems (MJSPSs) against stochastic deception attacks is discussed in this paper. Firstly, a novel dynamic event-based transmission protocol is provided to further decrease the proportion of sampled data into network. The effect of deception attacks is formulated as a random variable satisfying the Bernoulli distribution. And an asynchronous filter is delicately constructed. Based on the technique of linear matrix inequality (LMI), efficient criteria of stochastically stable for the filtering error systems with a predetermined dissipative performance are obtained. An effective method of jointly design the proposed dynamic event-triggered transmission protocol and the non-synchronous filter is offered. Lastly, a numerical instance and a resistance-capacitance (RC) circuit system are provided to display the effectiveness and the benefit of the developed method. •An dynamic event-triggered scheme is given to decrease the transmission rate.•The deception attacks are formulated as the Bernoulli distribution random variable.•The effect of asynchronous between MJSPSs and the filter is formulated as an HMM.•Based on the technique of LMI, criteria of stochastically stable are obtained.•The co-design of resilient filter and dynamic event-triggered scheme is obtained.