Abstract
We provide sufficient conditions for the existence, uniqueness and global Mittag-Leffler stability for the solutions of fractional difference model of bidirectional associative memory (BAM) neural networks. Before proceeding to the main results, we introduce the newly established discrete fractional calculus and expose some of its features. The techniques of Lyapunov function and Lyapunov direct method are then employed to prove the main results. We give two examples to verify and illustrate the theory.