Global exponential stability of high-order bidirectional associative memory (BAM) neural networks with time delays in leakage terms

Abstract

High-order bidirectional associative memory (BAM) neural networks with time delays in leakage terms play an increasingly important role in the design and implementation of neural network systems. Based on Lyapunov–Krasovskii functional and linear matrix inequality technique, issues of global exponential stability for high-order BAM neural networks with time delays in leakage terms are investigated. The proposed results, which do not require the boundedness, differentiability and monotonicity of the activation functions. In addition, the stability criteria that depend on the leakage time delays and our results are presented in terms of LMIs, which can be efficiently solved via a standard numerical package. Three numerical examples are provided to demonstrate the effectiveness of the proposed approach.