Non-fragile exponential synchronization of stochastic neural networks via impulsive control: An aperiodic intermittent technique

Abstract This study aims to address the non-fragile exponential synchronization problem of stochastic neural networks (SNNs). To cut down unnecessary control costs, a novel aperiodic intermittent-based impulsive control (APIIC) is designed for the first time in this investigation. Besides, the randomly occurring gain fluctuation (ROGF) is considered in APIIC, which satisfies certain Bernoulli distributed white noise sequences. By exploiting the Lyapunov approach, some sufficient criteria are derived in terms of linear matrix inequalities, which imply that APIIC can achieve exponential synchronization of SNNs with and without ROGF. More intriguingly, a technical definition of aperiodic windows-based average impulsive interval is developed to cut back the conservativeness of these results, meanwhile, these results take into account both synchronizing impulses and desynchronizing impulses. At last, the effectiveness of our explored results is confirmed by several numerical examples.