Event-based asynchronous and resilient filtering for singular Markov jump LPV systems against deception attacks

Abstract

This article addresses the issue of dissipative asynchronous and resilient filter design for singular Markov jump linear parameter-varying (SMJLPV) systems against deception attacks under the dynamic event-triggered transmission protocol. Firstly, an improved dynamic event-triggered transmission protocol is provided to further relieve the channel congestion caused by the bandwidth limited communication network. Then, the deception attack, which can potentially destroy the integrity of the system, is modelled as a random variable satisfying Bernoulli distribution. Since the filter cannot identify the original system mode accurately, the hidden-Markov-model (HMM) is established to describe the phenomenon that two modes are out of synchronization. By augmenting the states of the original system and the filter, the filtering error systems are converted into SMJLPV systems with partially known transition rates (TRs). Based on the parameter-dependent linear matrix inequalities (PDLMIs), a cooperative design technique for the asynchronous filter and the weighting matrix of the dynamic event-triggered transmission protocol is proposed. Lastly, a numerical instance and a resistance-inductance-capacitance (RLC) switch circuit system are employed to verify the effectiveness of the theoretical results.