Boolean Neural Networks

In this paper, we present a new type of neuron, called Boolean neuron. Further, we suggest the general structure of a neural network that includes only Boolean neurons and may realize several sets of Boolean functions. The advantages of these neural networks consist in the reduction of memory space and computation time in comparison to the representation of Boolean functions by usual neural networks. The Boolean neural network may be mapped to a FPGA so that our new training algorithms substitute classical design methods of these circuits. In the example, we show the decomposition of a set of Boolean functions into common basic functions and their mapping to the general Boolean neural network. Key-Words: Boolean neuron, BNN, data representation, FTFS, functions set