Robust stability of neural networks with discontinuous activation functions and time-varying delays

In this paper, dropped the assumption of the boundedness of the activation functions, the global dynamics are investigated for the recurrently connected neural networks (RCNNs) with discontinuous activations and time-varying delays. Based on the nonsmooth analysis theory, linear matrix inequality (LMI) technique and differential inclusions approach, several sufficient conditions are obtained to ensure the existence, uniqueness and global robust stability of the equilibrium point for the RCNNs. The obtained conditions are derived in terms of LMIs which are dependent on the size of the time-varying delay and the size of the time derivative of the time-varying delay. Finally, simulation examples are constructed to justify the proposed theoretical analysis.