Finite time anti-synchronization of complex-valued neural networks with bounded asynchronous time-varying delays

Xiwei Liu,Zihan Li
DOI: https://doi.org/10.1016/j.neucom.2020.01.035
IF: 6
2020-04-01
Neurocomputing
Abstract:<p>The finite time anti-synchronization of master-slave coupled complex-valued neural networks (CVNNs) with bounded asynchronous time-varying delays is studied. With the decomposing technique and generalized {<em>ξ</em>, ∞}-norm, several criteria for ensuring the finite time anti-synchronization are obtained. The whole anti-synchronized process is divided into two phases (it is called Two-Phases-Method, 2PM): first, the norm of each error state component will flow from initial value to 1 in finite time, then flow from 1 to 0 also in finite time, thus the whole processing time is finite. Finally, a numerical example is presented to demonstrate the effectiveness of our obtained results.</p>
computer science, artificial intelligence
What problem does this paper attempt to address?