(V)TEAM for SPICE Simulation of Memristive Devices With Improved Numerical Performance

Dalibor Biolek,Viera Biolkova,Zdenek Biolek,Shahar Kvatinsky,Zdenek Kolka

doi:10.1109/access.2021.3059241

Abstract

The paper introduces a set of models of memristive devices for a reliable, accurate and fast analysis of large networks in the SPICE (Simulation Program with Integrated Circuit Emphasis) environment. The modeling starts from the recently introduced TEAM (ThrEshold Adaptive Memristor Model) and VTEAM (Voltage ThrEshold Adaptive Memristor Model). A number of improvements are made towards the stick effect elimination and other numerical refinements to make the analysis of large networks fast and accurate. A set of models are proposed that utilize the synergy of several techniques such as window asymmetrization, integration with saturation, state equation preprocessing, scaling, and smoothing. The performance of models is tested in Cadence PSPICE 17.2 and particularly in HSPICE v2017, the latter on a large-scale CNN (Cellular Nonlinear Network) for detecting edges of binary images. The simulations manifest the usability of developed models for fast and reliable operation in networks containing more than one million nodes.

Highlights

In 2013, a model for current-controlled memristive devices, called TEAM (ThrEshold Adaptive Memristor Model), was published in [1]
The (V)TEAM falls into the category of phenomenological models, which are fitted to real memristive devices via tweaking a set of parameters
The results described below will make the universal models of memristive devices accessible to a wide community of SPICE users

Summary

INTRODUCTION

In 2013, a model for current-controlled memristive devices, called TEAM (ThrEshold Adaptive Memristor Model), was published in [1]. The corresponding extended memristors cannot be described via charge-flux constitutive relations, and the problem of natural versus physical state variable does not exist here It is obvious from the state equation (3) and from the diagram in Fig. 1 that the stick effect in unavoidable in (V)TEAM for an arbitrary symmetric window (4), which takes zero values at boundaries (such as the rectangular, Joglekar and Prodromakis window), so the method of state space transformation can play an important role in such cases. In combination with the given form of the differential equation (3) of VTEAM, it would imply an infinite increase in the state variable G, because the function g in the integral (14) evolves in time as asymmetric, with a nonzero DC component This finding points out the unsuitability of the given method for modeling memristors with window functions of type (5), which decrease to zero asymptotically. It is useful to compute the logarithms of such terms, which on principle cannot be zero (for example the window functions (5)), to sum them and perform the inverse exp operation

SMOOTHING

TESTING UNDER SINUSOIDAL EXCITATION The following 10 models were tested: R