Resilient 2-∞ filtering with dwell-time-based communication scheduling

作     者：Sun, Ying Ding, Derui Dong, Hongli Wei, Guoliang 

作者机构：Department of Control Science and Engineering University of Shanghai for Science and Technology Shanghai200093 China Institute of Complex Systems and Advanced Control Key Laboratory of Networking and Intelligent Control Northeast Petroleum University Daqing163318 China College of Science University of Shanghai for Science and Technology Shanghai200093 China 

出 版 物：《Nonlinear Analysis: Hybrid Systems》 (Nonlinear Anal. Hybrid Syst.)

年 卷 期：2020年第37卷

页      面：100901-100901页

核心收录：

学科分类：0711[理学-系统科学] 0809[工学-电子科学与技术（可授工学、理学学位）] 08[工学] 0811[工学-控制科学与工程] 0701[理学-数学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：This work was supported in part by the National Natural Science Foundation of China under Grant 61973219   Grant 61933007   Grant 61873169 and Grant 61873058   in part by the Natural Science Foundation of Shanghai  China under Grant 18ZR1427000   and in part by the Natural Science Foundation of Heilongjiang Province of China under Grant ZD2019F001 

主　　题：Linear matrix inequalities 

摘      要：This paper is concerned with the resilient filtering problem for a class of discrete-time stochastic nonlinear systems with both time-varying delays and probabilistic distributed delays. In order to save the limited communication resource, stochastic communication protocols (SCPs) with multiple rules governed by a switching signal are employed to schedule the data transmission between the sensors and the filter for complying with different network scenarios and preserving the desired performance. An approach based on average dwell time (ADT) is utilized to establish the rule of SCPs for different scenarios. The purpose of the filtering problem is to design a resilient filter such that, under the regulation of SCPs, the dynamics of the filtering error is exponentially mean-square stable, and satisfies a prescribed disturbance attenuation level in a 2-∞ sense. A sufficient condition for the exponential mean-square stability is first derived and the 2-∞ performance is then guaranteed. In terms of certain linear matrix inequalities, the solvability of the addressed problem is discussed and the explicit expression of the desired resilient filter is also parameterized. Finally, a numerical example is provided to demonstrate the validity of the proposed design approach. © 2020 Elsevier Ltd