Abstract
This paper addresses the problem of event-triggered $H_\infty $ filter design for nonlinear systems under both hybrid attacks and sensor saturation. A novel event-triggered mechanism (ETM) is proposed to reduce the number of transmission data per unit time, thereby improving the network QoS and taking the control performance into account. Considering that the deception attack may deteriorate the control performance, ETM is designed to be sensitive to deception attacks, that is, the average data release rate during the deception attack is higher than other periods. Consequently, the control performance can be improved. Moreover, the proposed ETM can reduce the occurrence of erroneous triggering events that are aroused by abnormal abrupt variation. In addition, the denial-of-service (DoS) attack is considered as well. A switching filtering error model is established based on the fact of each period of DoS attack being classified as active period and sleeping period. In addition, the problem of measurement saturation is concerned in filter design. By using Lyapunov-Krasovskii stability theory, sufficient conditions are obtained to guarantee the stability of the fuzzy filtering error system. The effectiveness of filter design methods is finally verified by a simulation example.
Highlights
Nonlinear systems are commonly and widely exist in the real-world [1]–[3], such as vehicle systems, social medical systems, etc
Remark 4: To enhance the estimation performance of the system against deception attacks, the item F (t) is introduced in the event-triggered mechanism (ETM) that we proposed in (9)
A novel ETM is adopted for releasing much more data compared when the system suffers from deception attacks compared to traditional ETM
Summary
Nonlinear systems are commonly and widely exist in the real-world [1]–[3], such as vehicle systems, social medical systems, etc. We deal with the problem of event-triggered H∞ secure filtering for T-S fuzzy-model-based nonlinear systems with sensor saturation and hybrid attacks. To get a prescribed estimation performance of nonlinear systems subject to DoS attacks and sensor saturation, a tolerant filtering design approach is proposed by converting the system into a switched T-S fuzzy-modelbased system. Based on such a model, a new resilient security criteria is put forward with consideration of the proposed ETM. An assumption to the membership function is made by θjg − ιjθjg 0 (0 < ιj ≤ 1)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
