Dynamic event-triggered distributed filtering design for interval type-2 fuzzy systems over sensor networks under deception attacks

Abstract

This article investigates the event-based distributed filtering design issue for interval type-2 (IT2) fuzzy systems over sensor networks under deception attacks. In sensor networks, while each sensor receives its measurement value, it can also sense neighbouring one's measurement based on the interconnected topology, thereby estimating system states. To reduce redundant transmission, a novel distributed dynamic event-triggered mechanism (DETM) is introduced to each sensor node. In particular, the proposed DETM contains an internal dynamic variable, which is a filtered version of the trigger event signal to increase the interval between two consecutive event triggers. Then, a new distributed filter error fuzzy system is well constructed under the consideration of DETM, networked-induced delay, and random deception attacks. To relax stability conditions, slack matrices with membership functions (MFs) information are used in the stability analysis. Furthermore, based on Lyapunov–Krasovskii functional approach, membership-functions-dependent sufficient criteria are obtained to ensure the established filter error fuzzy system is mean-square stability with performance. Two examples are finally shown to demonstrate the effectiveness of the filtering method.