State estimation for a class of artificial neural networks subject to mixed attacks: A set-membership method

This article deals with the set-membership state estimation problem for a class of artificial neural networks subject to time-delays and mixed malicious attacks. Both Denial-of-Service (DoS) and deception attacks are taken into consideration. The objective of the addressed problem is to design the state estimation algorithm for the artificial neural networks under investigation in spite of the existence of the malicious mixed attacks. By means of the set-membership approach in combination with certain convex optimization algorithm, the sufficient condition is established for the existence of the desired state estimator in terms of the solvability of a recursive matrix inequality. The resulting state estimation error is confined within certain pre-specified ellipsoidal region. An optimization problem is then formulated with the purpose of seeking the filtering parameters guaranteeing the locally optimal performance. Finally, the developed theoretical results are verified via an illustrative numerical example. (c) 2020 Elsevier B.V. All rights reserved.