Passive state estimation for Markov jumping inertial neural networks under fading channels

Abstract

SummaryThis article studies the issue of passive state estimation for inertial neural networks with Markov jumping parameters and fading channels. To facilitate the state estimator design, inertial neural networks which is initially described by the second‐order form, is converted into a first‐order one via a suitable variable substitution. Additionally, the channel fading problem, which is caused by the long‐distance propagation of the signal in the network environment, is described by Bernoulli sequences. Then, some sufficient criteria are established in view of the Lyapunov theory, which can guarantee that the estimation error system is stochastically stable and achieves the desired passive performance. A solution of the state estimator design problem for Markov jumping inertial neural networks with fading channels is obtained with the help of the criteria. Finally, an illustrative example is given to verify the effectiveness and validity of the obtained method.