A Dual-Read SOT-MRAM Bitcell Architecture for Robust In-Memory Computing

Spin-Orbit Torque Magnetoresistive RAM (SOT-MRAM) offers several advantages over traditional SRAM for in-memory computing (IMC), including non-volatility and lower leakage power. However, conventional SOT-MRAM-based IMC architectures suffer from reduced read margins due to the accumulation of bitcell output charges on shared bitlines during logic operations, which degrades robustness and sensing accuracy. In this work, we propose a novel 2T-1R SOT-MRAM bitcell architecture featuring a dual-transistor per write-path topology, enabling independent and simultaneous reads from two separate bitcells, each with its own peripheral sense amplifier. This dual-read approach significantly enhances the sense margin and reduces cross-talk compared to conventional architectures that rely on shared read paths. Furthermore, the proposed design enables robust in-memory Boolean operations—such as NAND, NOR, and XOR—positioning it as a promising solution for reliable IMC systems.

Authors

Institutions

Publication Details

Published
2025-08-04
DOI
https://doi.org/10.1109/coins65080.2025.11125780
Primary Topic
Advanced Memory and Neural Computing
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

A Dual-Read SOT-MRAM Bitcell Architecture for Robust In-Memory Computing

Akhilesh Jaiswal, Rouhan Noor, Md Abdullah-Al Kaiser
1 links
Advanced Memory and Neural Computing
article

A Dual-Read SOT-MRAM Bitcell Architecture for Robust In-Memory Computing

Akhilesh Jaiswal, Rouhan Noor, Md Abdullah-Al Kaiser
article en

Abstract

Spin-Orbit Torque Magnetoresistive RAM (SOT-MRAM) offers several advantages over traditional SRAM for in-memory computing (IMC), including non-volatility and lower leakage power. However, conventional SOT-MRAM-based IMC architectures suffer from reduced read margins due to the accumulation of bitcell output charges on shared bitlines during logic operations, which degrades robustness and sensing accuracy. In this work, we propose a novel 2T-1R SOT-MRAM bitcell architecture featuring a dual-transistor per write-path topology, enabling independent and simultaneous reads from two separate bitcells, each with its own peripheral sense amplifier. This dual-read approach significantly enhances the sense margin and reduces cross-talk compared to conventional architectures that rely on shared read paths. Furthermore, the proposed design enables robust in-memory Boolean operations—such as NAND, NOR, and XOR—positioning it as a promising solution for reliable IMC systems.

University of Wisconsin–Madison (US)
Affordable and clean energy
Openalex Percentile: Top 79%
Advanced Memory and Neural Computing
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

A Dual-Read SOT-MRAM Bitcell Architecture for Robust In-Memory Computing — Akhilesh Jaiswal, Rouhan Noor, et al. · (2025) | TGRS Research Map | TGRS