Artificial neuron structure and preparation method thereof, and signal and time extraction method

Abstract

The invention provides an artificial neuron structure and a preparation method thereof, and a signal and time extraction method. The artificial neuron structure comprises a substrate, a back gate metal layer, an epitaxial layer, two non-contact source drain metal layers, and an organic thin film layer, wherein the back gate metal layer is positioned below the substrate; the epitaxial layer is positioned above the substrate; the two non-contact source drain metal layers are positioned above the epitaxial layer; and the organic thin film layer is in contact with the two source drain metal layersseparately and are stacked on the epitaxial layer, wherein two holes are formed in the organic thin film layer for exposing at least one part of the two source drain metal layers. According to the artificial neuron structure, by simulating the working principle of the biological neuron, the carrier condensation distribution in a semiconductor channel is regulated and controlled by charges provided by an electrochemical reaction to realize corresponding output for multiple kinds of electrical stimulation; and the artificial neuron structure has the time-dependent change characteristic, can perform power-down operation and can reduce static state power consumption to be zero, is suitable for application of a low-power-consumption circuit, and has wide application prospect.

Description

technical field [0001] The invention relates to the field of semiconductor device technology and artificial intelligence, in particular to an artificial neuron structure, a preparation method thereof, and a signal and time extraction method. Background technique [0002] Artificial neural network based on neural network algorithm combined with high performance computing is a hot spot in the field of artificial intelligence, and it is also a technological highland that major technology companies are competing for. In the above artificial neural network, high-performance computing is usually mainly accomplished by using a commercial high-performance central processing unit (CPU) and a graphics processing unit (GPU) combined with advanced neural network algorithms. Compared with the real biological neural network, the above artificial neural network has obvious speed advantages, but there is still a big gap in power consumption and integration. Therefore, not only simulating n...

AI Technical Summary

Problems solved by technology

The above devices or circuits have a common problem: the static power consumption is relatively large

However, the introduction of the above-mentioned new materials and new structures has limited reduction in device power consumption.

Images