Memristor-based vector neural network architecture  

作　　者：Hai-Jun Liu Chang-Lin Chen Xi Zhu Sheng-Yang Sun Qing-Jiang Li Zhi-Wei Li 刘海军;陈长林;朱熙;孙盛阳;李清江;李智炜(College of Electronic Science,National University of Defense Technology,Changsha 410073,China)

机构地区：[1]College of Electronic Science,National University of Defense Technology,Changsha 410073,China

出　　处：《Chinese Physics B》2020年第2期463-467,共5页中国物理B（英文版）

基　　金：Project supported by the National Natural Science Foundation of China(Grant Nos.61471377,61804181,61604177,and 61704191).

摘　　要：Vector neural network(VNN)is one of the most important methods to process interval data.However,the VNN,which contains a great number of multiply-accumulate(MAC)operations,often adopts pure numerical calculation method,and thus is difficult to be miniaturized for the embedded applications.In this paper,we propose a memristor based vector-type backpropagation(MVTBP)architecture which utilizes memristive arrays to accelerate the MAC operations of interval data.Owing to the unique brain-like synaptic characteristics of memristive devices,e.g.,small size,low power consumption,and high integration density,the proposed architecture can be implemented with low area and power consumption cost and easily applied to embedded systems.The simulation results indicate that the proposed architecture has better identification performance and noise tolerance.When the device precision is 6 bits and the error deviation level(EDL)is 20%,the proposed architecture can achieve an identification rate,which is about 92%higher than that for interval-value testing sample and 81%higher than that for scalar-value testing sample.Vector neural network(VNN) is one of the most important methods to process interval data.However,the VNN,which contains a great number of multiply-accumulate(MAC) operations,often adopts pure numerical calculation method,and thus is difficult to be miniaturized for the embedded applications.In this paper,we propose a memristor based vector-type backpropagation(MVTBP) architecture which utilizes memristive arrays to accelerate the MAC operations of interval data.Owing to the unique brain-like synaptic characteristics of memristive devices,e.g.,small size,low power consumption,and high integration density,the proposed architecture can be implemented with low area and power consumption cost and easily applied to embedded systems.The simulation results indicate that the proposed architecture has better identification performance and noise tolerance.When the device precision is 6 bits and the error deviation level(EDL) is 20%,the proposed architecture can achieve an identification rate,which is about 92% higher than that for interval-value testing sample and 81% higher than that for scalar-value testing sample.

关 键 词：MEMRISTOR memristive DEVICES VECTOR NEURAL NETWORK INTERVAL 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]