LMI conditions for some dynamical behaviors of fractional-order quaternion-valued neural networks

Abstract

Abstract This paper addresses the issue of three dynamical behaviors including global Mittag-Leffler stability, robust stability and projection synchronization for fractional-order quaternion-valued neural networks (FQVNNs). Some linear matrix inequality conditions for these dynamical behaviors of FQVNNs are given by Lyapunov stability theory, quaternion matrix theory, Homeomorphic mapping theory and fractional differential equation theory. Furthermore, these obtained sufficient conditions for stability and synchronization are superior to those in existing literature. Finally, three examples are given to illustrate the effectiveness of the theoretical results.

Keywords

• Fractional-order quaternion-valued neural networks
• Global Mittag-Leffler stability
• Projective synchronization
• Robust stability
• Linear matrix inequality