Nonfragile l2-l∞ state estimation for discrete-time neural networks with jumping saturations

• A more general saturation model is proposed, which changes with the environment.
• The nonfragile estimator is designed to overcome the randomly occurring disturbances around the estimator to improve its robustness.
• Sufficient conditions are established to ensure that the estimation error system is stochastically stable and satisfies l2l2–l∞l∞ performance.

In this paper, the nonfragile l2−l∞ state estimation problem is investigated for the neural networks with sensor saturations. In order to model the phenomenon that the sensor saturation varies with the environment, a multi-saturation model is proposed and a homogenous Markov chain is introduced to describe the variation. The nonfragile state estimator which can be used to improve the robustness of the estimator with randomly occurring uncertainty is introduced. Sufficient conditions are established to ensure that the estimation error system is stochastically stable and satisfies l2l2–l∞l∞ performance, the estimator gains are derived via solving the linear matrix inequalities. Finally, an example is provided to illustrate the effectiveness of the proposed new design techniques.