Exponential Asynchronous Stabilization for Delayed Semi-Markovian Neural Networks via DAEIC

The exponential asynchronous stabilization (EAS) issue for a category of neural networks (NNs) with semi-Markov jump (SMJ) parameters and additive time-varying delays (ATDs) is addressed in this article. Here, the SMJ parameters in the controller gain are supposed to be distinct from those in the system structure, which is more consistent with the actual situation. To further relieve the communication load of the network, a new discrete adaptive event-triggered impulsive control (DAEIC) scheme is proposed, where the impulsive moments are the sampling instants satisfying event-triggered constraints, and the triggering threshold can be dynamically adjusted by an adaptive update rule (AUR) related to the current sampling state and the last triggered state. A more flexible looped Lyapunov-Krasovski functional (LLKF) is constructed to commendably capture the available information about impulsive instants, triggering state, sampling interval, ATDs, and heterogeneous SMJ parameters. Combined with the LLKF, DAEIC scheme, and other inequality analysis approaches, some novel results guaranteeing the EAS of the underlying systems are exported. Finally, three explanatory examples are presented to check the validity of our results.