随着对计算机性能要求的不断提高,人们一直在寻找能像人脑一样具有学习记忆功能的新型计算机。自从2008年惠普实验室发现忆阻器以后,发展具有人脑水平的智能计算机成为可能。众所周知,突触是大脑神经网络的基本单元,突触可塑性是学习和记忆的生物学基础。因此,为了实现具有学习和记忆功能的智能计算机,利用忆阻器模拟突触可塑性至关重要。综述了忆阻器在模拟突触的增强、抑制、短时程可塑性和长时程可塑性方面的研究现状,并对其研究前景进行了展望。
New concept computers working like brain have long been sought along with increased requirements for the computer performance.It is possible to develop intelligent computer towards the level of the brain since mem-ristor was found at HP laboratory in 2008.It is generally known that synapse is the basic unit of the brain neural net-work,and synaptic plasticity is regarded as biological basis of learning and memory.Thus,in order to realize the in-telligent computer with functions of learning and memory,it is important to simulating synaptic plasticity based on memristor.The research situation of memristor on synaptic simulation,including synaptic potentiation,depression, short-term plasticity,long-term plasticity as well as the research development in the field are reviewed.