期刊名称：Balkan Journal of Electrical & Computer Engineering
印刷版ISSN：2147-284X
出版年度：2019
卷号：7
期号：1
页码：34-43
DOI：10.17694/bajece.457902
语种：English
出版社：Kirklareli University
摘要：Artificial neural networks are successfully used for classification,prediction, estimation, modeling and system control. However, artificial neuralnetworks integrated circuits are expensive and not matured enough. Memristorsor memristive systems which show a nonvolatile memory behavior has a highpotential for use in artificial neural network circuit applications. Somememristive synapse or memristive neural network applications already exist inliterature. The complementary memristor or resistive switch memories have beensuggested as an alternative to one-cell memristor memories. Their sensing ismore difficult and complex than the others. The complementary memristor memorytopologies with a sensing node are also inspected in literature. To the best ofour knowledge, a neural network circuit which is based on the complementaryresistive switches with a sensing/writing node does not exist in literatureyet. In this paper, several neuralnetwork circuits which are based on the complementary resistive switches with asensing/writing node have been designed and examined for the first time inliterature. Their analysis are given and simulations are performed to verifytheir operation. We expect that such a complementary resistive switchimplementation may find use in artificial neural networks chips in the future.
