A practical hafnium-oxide memristor model suitable for circuit design and simulation

A New Simplified Model and Parameter Estimations for a HfO2-Based Memristor †

“…Equations (1) and (2) fully define the corresponding applied physics-based memristor model [7,8]. The ionic transport is linked to the ionic dopant drift and the electron motion in the respective hafnium dioxide material [10][11][12]. The applied window function is a modification of the standard Biolek window function [4]:…”

“…This memristor model was applied after a tuning procedure according to the experimental current-voltage relationships of an HfO 2 memristor element [10] in the present investigation [7,8]. The main interest in the present case is associated with new and not very well analyzed types of memristors, such as transition-metal oxide memristors, especially hafnium oxide memristors [10][11][12]. Several basic models of HfO 2 -based memristors exist in the technical literature [10][11][12][13].…”

“…The main interest in the present case is associated with new and not very well analyzed types of memristors, such as transition-metal oxide memristors, especially hafnium oxide memristors [10][11][12]. Several basic models of HfO 2 -based memristors exist in the technical literature [10][11][12][13]. To the best of the author's knowledge, some of them are either very complex [10] or not sufficiently precise [11].…”

“…Several basic models of HfO 2 -based memristors exist in the technical literature [10][11][12][13]. To the best of the author's knowledge, some of them are either very complex [10] or not sufficiently precise [11]. Other models, such as Biolek's [4] and Joglekar's [3], are not able to correctly express the corresponding i-v relationship due to the low nonlinearity of the current-voltage characteristic.…”

“…Other models, such as Biolek's [4] and Joglekar's [3], are not able to correctly express the corresponding i-v relationship due to the low nonlinearity of the current-voltage characteristic. The motivation for the present investigation was to offer a simple nonlinear ionic drift model for a HfO 2 memristor with a strongly nonlinear window function [14,15],parameter estimations of the model using the least squares algorithm in MATLAB-Simulink [15,16], to minimize the root mean square error and provide a simulation the memristor model with the obtained parameters, to compare the derived errors with those obtained using the applied methodology, to compare the errors with those derived using the best existing memristor models [10,11], and to replace the step function in the considered model using a differentiable tangent hyperbolic function [17]. For this aim, the tunable modified nonlinear memristor model proposed in References [7,8] is very suitable, in combination with the highly nonlinear window function, offered by Mladenov and Kirilov [14].…”