Stability Analysis for Discrete-Time Switched Positive Neural Networks with Mixed Delays

This paper studies the asymptotical stability for a class of discrete-time switched positive neural networks with mixed delays. The time-varying delays and distribution delays which exist during neuron transmission are taken into consideration. The nonnegative state space of the first quadrant is divided into several subspaces, where the overlapping region is constructed to deal with the chattering behavior caused by frequent switching. Then hysteresis switching rule is designed by using the single copositive Lyapunov function (SCLF) and multiple copositive Lyapunov functions (MCLFs) methods, respectively. The sufficient conditions for the system's stability are given in the form of linear inequalities. At the end of this paper, two numerical examples demonstrate the effectiveness of this approach.