Hardware circuit for simulation of pulse neurons based on MOS field effect transistor and application thereof

Abstract

The present invention discloses a hardware circuit for simulation of pulse neurons based on an MOS field effect transistor. The circuit is formed by 8 MOS transistors and 2 simulation film capacitors,wherein a g pole and a d pole of a fifth MOS transistor are correspondingly connected with a g pole and a d pole of a sixth MOS transistor; a g pole and a d pole of a seventh MOS transistor are correspondingly connected with a g pole and a d pole of an eighth MOS transistor; a d pole of a first MOS transistor is connected with an s pole of a second MOS transistor, the g poles of the first MOS transistor, the second MOS transistor and the fourth MOS transistor are connected with a b end of a second end of a second simulation film capacitor, d poles of the second MOS transistor and the third MOS transistor and a ends of first ends of a first simulation film capacitor and the second simulation film capacitor are connected; a b end of a second end of the first simulation film capacitor and spoles of the four MOS transistor, the sixth MOS transistor and the eighth MOS transistor are earthed. The hardware circuit can simulate the functions of neurons, has the characteristics of a simple structure, high integration, low power consumption and easy maintenance. The hardware circuit can be applied in the simulation neural network.

Description

technical field [0001] The invention relates to artificial intelligence technology, in particular to a hardware circuit for simulating pulse neurons based on MOS field effect transistors and its application. Background technique [0002] Artificial intelligence has become one of the most concerned and fastest-growing technical fields in the world today. One of its most important features is to realize the intelligence of electronic devices or systems to improve the efficiency of information processing and storage. [0003] The rapid development of neuroscience has provided a new development direction for the field of artificial intelligence. Researchers proposed to improve the efficiency of calculation and storage by simulating biological neurons to transmit and calculate information with pulse signals, and thus proposed A spiking neural network model that mimics the neural network of the brain. At present, the operation method of spiking neural network has been applied in ...

AI Technical Summary

Problems solved by technology

[0004] At present, many researchers have proposed different neuron circuit models, but there are still problems such as complex circuit structure, low integration, high power consumption, and inconvenient maintenance.

Images