SINGLE-LAYER LINEAR NEURAL NETWORK EMPLOYING CELL SYNAPSE STRUCTURE

摘要

A single-layer linear neural network employing a cell synapse structure comprises a presynapse and a postsynapse. The presynapse comprises m pre-stage resistors. Respective ends of the m pre-stage resistors in the presynapse are all connected to an intermediate point, and respective other ends thereof are respectively connected to m pre-stage signal input ends. The pre-stage signal input end is used to receive an input voltage. The postsynapse comprises n post-stage resistors. Respective ends of the n post-stage resistors in the postsynapse are all connected to the intermediate point, and respective other ends are respectively connected to n post-stage signal output ends. The post-stage signal output end is used to output a current. In the single-layer linear neural network, the number of resistors is reduced. In addition, only two variable resistors or one of the variable resistors needs to be adjusted in order to change weights between external presynaptic neurons and postsynaptic neurons.