Extended dissipative state estimation for neural networks with semi‐Markov jump parameters and sensor nonlinearities against injection attacks
Year: 2025
Type: article
Abstract: This work focuses on developing a mode‐dependent estimator for neural networks that have semi‐Markov jump parameters and sensor nonlinearities under the imperfect network environment. Semi‐Markov jump process with its unique superiority is employed to depict mode switching within neural networks. The data sent over a communication network are considered to be at risk of cyber‐attacks, where adversaries could introduce false data in a probabilistic manner. Due to complex practical situations, the sensor may exhibit nonlinear characteristics. A state estimator is utilized to produce estimates for the underlying neural networks, incorporating the sensor's nonlinear characteristics. Furthermore, an applicable estimator is devised by addressing the convex optimization problem. Lastly, the efficacy of the formulated estimator is demonstrated through a numerical illustration.
Source: Asian Journal of Control
DOI: 10.1002/asjc.3668
Authors: Yuan Wang, Huaicheng Yan, Xiao Hu, Yufang Chang, Chaoyang Chen
Institutions: East China University of Science and Technology, Suzhou University of Science and Technology, Key Laboratory of Smart Manufacturing in Energy Chemical Process, Ministry of Education, Hubei University of Technology, Xi'an University of Technology, Hunan University of Science and Technology
Language: English
Subfield: Control and Systems Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4409404671