An Event-Triggered Method for Stabilization of Stochastic Quaternion-Valued Memristive Neural Networks

The stochastic disturbances are common in real world and usually cause significant influence to engineering system. In this work, the stochastic disturbance is introduced into the quaternion-valued memristive neural networks (QVMNNs). The exponential input-to-state stabilization (EITSS) problem of stochastic QVMNNs is investigated. In order to be more effective and less costly in real applications, an event-triggered control strategy is adopted. The original QVMNNs are separated into four equivalent real-valued NNs by using Hamilton rule. Then, by using the Lyapunov functional approach and stochastic analysis technique, novel sufficient conditions for mean square EITSS of stochastic QVMNNs are derived. Moreover, it is proved that Zeno behavior will not take place in our event-triggered control method. Thus, the mean square EITSS problem of stochastic QVMNNs is solved in this work with less control cost. Lastly, simulation is performed to manifest the correctness of the theorem.