Metal Oxide Neural Devices and Their Applications

Abstract

Metallic oxide neurons are a potential candidate for future applications in neuroscience and neuroscience, where they can be used as a model for simulating synaptic functions in biology. This paper mainly introduces two kinds of oxide-based memristors: memristor and neural transistor, as well as briefly discuss their integration. Because of the two devices’ simple structure, similar to synaptic structure, high efficiency, low power consumption, high-density integration, compatibility with CMOS process conditions, and continuous control of conductance, it is considered to be the first choice for brain-like computing hardware. And the memristor can realize the integration of computing and storage functions in a single device. However, some problems and challenges still exist in memristor application. In the process of the device, there is a problem of insufficient miniaturization, because of the influence of leakage current, which limits the integration of the array and increases the operation energy consumption. In conclusion, despite their significant barriers before becoming commercially viable, their promising future is certain.