Neuronal circuits and neuromorphic circuits

Abstract

The present invention relates to a neuron circuit and a neuromorphic circuit, the neuron circuit includes a pulse generating circuit and a U unit connected to the control terminal of the pulse generating circuit; the pulse generating circuit is used to convert the input current into a pulse sequence , the U unit is used to control the excitation mode and pulse interval of the pulse generated by the pulse generating circuit. The neuron circuit and the neuromorphic circuit provided by the present invention use the U unit to control the pulse generation circuit to achieve different biological characteristics without using any bias voltage. The method realizes different biological characteristics, avoids the sensitivity of the circuit itself to the fluctuation of process, voltage and temperature, and makes the circuit easier to realize; at the same time, there is no normally open path in the circuit, so that the circuit consumes less power.

Description

technical field [0001] The invention relates to the technical field of neural networks, in particular to a neuron circuit and a neuromorphic circuit. Background technique [0002] As the scale of artificial neural networks increases, the area and power consumption issues become more and more important. Therefore, the main problem to be considered in the design of neuronal circuits today is how to achieve the richest possible biological properties at the lowest possible cost. [0003] The initial neuron circuit is often implemented using the Hodgkin-Huxley model. Although this model can describe biological behavior most accurately, the complex and huge parameters and formulas make it difficult to implement in hardware. To solve this problem, people began to look for other neuron models, including the accumulation and release (integrated & fire) model. The integrated & fire model is widely used because of its low cost of implementation, but its insufficient biological proper...

AI Technical Summary

Problems solved by technology

[0004] However, in the existing technology, no matter what method is used to build neuron circuits, different biological characteristics are often achieved by changing its own parameters or adjusting the bias voltage, which makes the circuit itself sensitive to fluctuations in process, voltage and temperature. quite sensitive

Not only that, in the prior art, neuron circuits often have a path that is always on, which in turn causes additional power consumption

Images