NOISE ISSUES IN HOLOGRAPHIC PHOTOREFRACTIVE INTERCONNECTIONS - APPLICATION TO NEURAL NETWORKS

We investigated the influence of the imperfections of some optical com ponents in a reconfigurable holographic interconnection system. These interconnections are recorded in a photorefractive crystal by interfer ing two light beams modulated by two spatial light modulators, We expe rimentally show that such set-ups are limited to small interconnect ca pacities by the finite contrast of the spatial light modulators. We pr oposed two techniques to improve the performances of such interconnect ion systems. We used this set-up to implement the first layer of a two layer neural network. With this system, we successfully recognised 10 1 chinese ideograms. (C) 1997 Elsevier Science B.V.