EIGENSTRUCTURE BIDIRECTIONAL ASSOCIATIVE MEMORY - AN EFFECTIVE SYNTHESIS PROCEDURE

In the present letter we propose a computationally efficient synthesis procedure for a class of bidirectional associative memories. Networks are described by a system of first-order ordinary difference equation s which are defined on a closed hypercube of the state-space with solu tions extended to the corner of the hypercube. The proposed algorithm possesses several advantages, since it is possible to exert control ov er the number of spurious states, since it is possible to estimate the basins of attraction of the stable memories, and since it is possible , under certain constraints, to effectively store a number of desired stable memories which by far exceed the order of the network. The appl icability of the present results is demonstrated by means of a specifi c application in reconfigurable control of flexible structures.