Optimal bidirectional associative memories

This paper proposes a model for linear optimal bidirectional associative me mories (BAMs), in which the design of the weight matrix is based on a least -squares cost functional. The minimization of the functional leads to a Lya punov-type matrix equation. The mathematical properties of such an optimal bidirectional system are studied in detail. Simulations are conducted for t he bidirectional association of pattern pairs with grey levels. It is shown that this optimal design of BAM is effective for memorizing patterns with grey levels.