Off-campus WSU users: To download campus access dissertations, please use the following link to log into our proxy server with your WSU access ID and password, then click the "Off-campus Download" button below.

Non-WSU users: Please talk to your librarian about requesting this thesis through interlibrary loan.

Access Type

WSU Access

Date of Award

January 2024

Degree Type

Thesis

Degree Name

M.S.

Department

Electrical and Computer Engineering

First Advisor

Gozde Tutuncuoglu

Abstract

Memristors are a type of non-volatile memory that possess several advantageous properties, including low latency, small device footprint, fast switching speeds, and low energyconsumption. These properties make them ideal candidates for implementation in neuromorphic hardware and analog accelerators. However, state-of-the-art devices are likely to suffer from critical performance degradation and device performance non-idealities. Key issues include nonlinear and asymmetric conductance tuning, along with device-to-device (D2D) and cycle-to-cycle (C2C) variability, potentially causing weight encoding and inference errors. In this work, I present a robust and novel characterization methodology to comprehensively analyze performance metrics, C2C, and D2D variability in commercial self-directed channel (SDC) memristors. I also present the crossbar array circuitry design based on the VTEAM memristor model. This array can accept inputs in the form of a series of pulses, apply a VMM operation, and convert the output back to voltage.

Off-campus Download

Share

COinS