Work

Reliable multi-level cell programming in FeFET arrays for in-memory computing

Year: 2025

Type: article

Abstract: With the increasing demands of AI and edge computing, traditional computing architectures face challenges in efficiency and scalability. This paper explores the use of multi-level cell FeFET arrays for in-memory computing systems, highlighting a program/verify scheme to precisely control threshold voltages (VT). Our study demonstrates the ability of a 1k FeFET crossbar to achieve up to seven distinct VT states, allowing nearly 3-bit storage per cell. This approach maintains computational accuracy and enhances storage density, offering a viable solution for energy-efficient, high-performance memory architectures.

Source: Japanese Journal of Applied Physics

DOI: 10.35848/1347-4065/adc745

Authors: Alptekin Vardar, Marcel Günther, Franz Müller, Nellie Laleni, Konrad Seidel, Thomas Kämpfe

Institutions: Fraunhofer Institute for Photonic Microsystems, Center Nanoelectronic Technologies

Language: English

Topic: Advanced Memory and Neural Computing

Subfield: Electrical and Electronic Engineering

Field: Engineering

Domain: Physical Sciences

Open Access status: hybrid

Estimated APC paid: $2,210

OpenAlex ID: W4409013347

FWCI: 0.907

Cites: 31

Cited by: 4

Related to: 10