Frequency-coded control of the conductance of memristors based on nanoscale layers of LiNbO$_3$ and (Co$_{40}$Fe$_{40}$B$_{20}$)$_{x}$(LiNbO$_{3}$)$_{100-x}$ composite in trained spiking neural networks
Abstract:
The memristive properties of Cu/nanocomposite/LiNbO$_3$/Cu capacitor structures based on a (Co$_{40}$Fe$_{40}$B$_{20}$)$_{x}$(LiNbO$_{3}$)$_{100-x}$ nanocomposite and an amorphous LiNbO$_3$ interlayer with thicknesses of about 40 and 20 nm, respectively, have been studied. It was found that these structures have relatively low resistive switching voltages ($\sim$2 V) and are capable of withstanding more than 10$^4$ cyclic switchings due to the formation of conducting channels in LiNbO$_3$ in fixed regions specified by the position of percolation chains of CoFe nanograins in the nanocomposite. It is shown that the conductance of Cu/nanocomposite/LiNbO$_3$/Cu memristors can vary according to local biosimilar rules. A simple neural network based on such memristors, trained by feeding a frequency-coded noise signal to its inputs, was implemented.