Dissipativity and Stability for Stochastic Non-Integer-Order Memristive BAM System with Leakage Terms and Mixed Delays

This paper is concerned with the problems of mean-square global dissipativity and global asymptotic stability for a class of stochastic fractional-order memristive BAM neural networks with leakage terms and mixed time-varying delays, including discrete delays and distributed delays. By using differential inclusion theory, stochastic analysis, matrix measure approach, and Lyapunov stability theory combined with linear matrix inequalities (LMIs), several new sufficient conditions are derived to ensure the mean-square global dissipativity and global asymptotic stability of the considered system. Compared with the existing results, the obtained stability and dissipativity criteria are less conservative due to the adoption of matrix measure and fractional-order differential inequalities. The proposed model simultaneously incorporates stochastic perturbations, memristive discontinuity, leakage effects, and mixed delays, which makes it more consistent with actual engineering scenarios such as pattern recognition and intelligent control. Finally, a numerical example is provided to demonstrate the effectiveness and correctness of the theoretical results.