Exponential stability on a class of delayed neural networks of neutral type

Consider of a class of neural networks of neutral type which have time-varying delay and parametric uncertainties, A sufficient condition is given to provide the uniqueness and exponential stability of the equilibrium point for this system by constructing a Lyapunov function, this method is independent of the amplitude of time delays and it doesn't have to assume the boundness, strict monotonicity and differentiability of neuron excitation function. this condition is only dependent on the interconnected matrices and derivative of time delays. the criterion can be expressed in terms of LMIs which is easy to deal with. Finally, a numerical example is given to illustrate the effectiveness and feasibility.