METHOD OF PROCESSING INFORMATION IN NEURAL NETWORKS Russian patent published in 2010 - IPC G06F1/00 

FIELD: information technology.

SUBSTANCE: in neural networks, digital data codes represented by electrical signals, which are converted in accordance with logic output inference with functional transformation of digital codes of coordinates of the input vector to values of grade of membership to two membership functions "small value" and "large value", are transmitted, scaled through a prod procedure, digital codes of weighted values are transmitted to inputs of logic formal neurons min and coordinates of an intermediate vector are formed, the digital code of each of which is converted to two values of digital codes and a first and a second intermediate vector is formed, digital codes of coordinates of the first of which are scaled through a prod procedure and transmitted to inputs of additional logic forma neurons min for formation of an intermediate vector, digital coordinates of the third and fourth intermediate vectors are scaled through a prod procedure, transmitted to inputs of output logic formal neurons max and digital codes of coordinates of the output vector of the neural network are formed.

EFFECT: higher functional stability and information security in neuro-fuzzy networks owing to increase in information redundancy.

1 dwg