HMM-based finite-time synchronization of fuzzy jumping neural networks with input constraints and partial information

This paper investigates the hidden Markov model-based finite-time synchronization problem for discrete-time fuzzy jumping neural networks with input constraints. To cope with the fact that obtaining system mode information is difficult normally, a general case when either transition probabilities or observation probabilities of jumping processes are assumed to be partially known is taken into account. Furthermore, the input constraints which may lead to the failure of the presented designed method are also considered. Based on this, a finite-time synchronization criterion is established by using the observation signal, and an effective control scheme with less conservatism is given with the help of the activation function division method and hidden Markov model-based method. Finally, an example is used to demonstrate the effectiveness of the proposed method.