Standard cell-based implementation of a digital optoelectronic neural-network hardware

A standard cell-based implementation of a digital optoelectronic neural-net work architecture is presented. The overall structure of the multilayer per ceptron network that was used, the optoelectronic interconnection system be tween the layers, and all components required in each layer are defined. Th e design process from VHDL-based modeling from synthesis and partly automat ic placing and routing to the final editing of one layer of the circuit of the multilayer perceptrons are described. A suitable approach for the stand ard cell-based design of optoelectronic systems is presented, and shortcomi ngs of the design tool that was used are pointed out. The layout for the mi croelectronic circuit of one layer in a multilayer perceptron neural networ k with a performance potential 1 magnitude higher than neural networks that are purely electronic based has been successfully designed. (C) 2001 Optic al Society of America.