LMI-based criteria for stability of high-order neural networks with time-varying delay (Q867945)

The authors consider a class of high-order neural networks with time-varying delay of the form \[ x^{\prime}_i(t) = -a_i x_i(t) +{\sum}_{j=1}^Lw_{ij}{\Pi}_{k\in I_j}[g_k(x_k(t))]^{d_k(j)} + {\sum}_{j=1}^Lw_{ij}^{\tau}{\Pi}_{k\in I_j}[g_k(x_k(t-\tau (t)))]^{d_k(j)} \] for \(i=1,2,\dots, n\). Under appropriate assumptions this system has a unique equilibrium point. Additional sufficient conditions for the global exponential stability of the equilibrium point are formulated. Some numerical examples are given to illustrate the theoretical results.