High-Speed and Low-Cost In-Array Memristive Multipliers Using SIXOR and TMSL Logics

IF 2.1 4区 工程技术 Q3 ENGINEERING, ELECTRICAL & ELECTRONIC
Roya Rahimi Disfani;Mojtaba Valinataj;Nima TaheriNejad
{"title":"High-Speed and Low-Cost In-Array Memristive Multipliers Using SIXOR and TMSL Logics","authors":"Roya Rahimi Disfani;Mojtaba Valinataj;Nima TaheriNejad","doi":"10.1109/TNANO.2025.3566272","DOIUrl":null,"url":null,"abstract":"Memristive systems have many promising features, making them suitable for both storage and computation. Memristors can perform logical operations and they can be used as the basic structures in digital circuits such as adders and multipliers. In this paper, at first, a new fast and low-cost Full-Adder (FA) is proposed using Single-cycle In-memristor XOR (SIXOR) and Three Memristors Stateful Logic (TMSL) gates that benefits from the advantages of both logics. Then, the proposed FA is used as one of the basic units inside two new array multipliers. The first proposed multiplier is designed in such a way that it has the lowest computational steps (delay) among the existing designs. This design has on average around 70% lower delay compared to the existing designs. The second proposed multiplier, as the low-cost design, requires a very low number of memristors thanks to reusing the existing resources more efficiently, while still having a low delay. This multiplier achieves on average around 36% memristor reduction compared to the state-of-the-art multipliers. Based on the analysis, both proposed array multipliers have notable efficiency advantages compared to the state-of-the-art designs based on different Figures of Merit (FoMs). For example, based on the balanced Figure of Merit (FoM), in which the number of computational steps and the number of required memristors have equal weight, the first and the second proposed multipliers achieve up to 4.6× and 14.9× improvements, respectively, compared to the existing designs in 64-bit multiplication.","PeriodicalId":449,"journal":{"name":"IEEE Transactions on Nanotechnology","volume":"25 ","pages":"13-25"},"PeriodicalIF":2.1000,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Transactions on Nanotechnology","FirstCategoryId":"5","ListUrlMain":"https://ieeexplore.ieee.org/document/10981683/","RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}
引用次数: 0

Abstract

Memristive systems have many promising features, making them suitable for both storage and computation. Memristors can perform logical operations and they can be used as the basic structures in digital circuits such as adders and multipliers. In this paper, at first, a new fast and low-cost Full-Adder (FA) is proposed using Single-cycle In-memristor XOR (SIXOR) and Three Memristors Stateful Logic (TMSL) gates that benefits from the advantages of both logics. Then, the proposed FA is used as one of the basic units inside two new array multipliers. The first proposed multiplier is designed in such a way that it has the lowest computational steps (delay) among the existing designs. This design has on average around 70% lower delay compared to the existing designs. The second proposed multiplier, as the low-cost design, requires a very low number of memristors thanks to reusing the existing resources more efficiently, while still having a low delay. This multiplier achieves on average around 36% memristor reduction compared to the state-of-the-art multipliers. Based on the analysis, both proposed array multipliers have notable efficiency advantages compared to the state-of-the-art designs based on different Figures of Merit (FoMs). For example, based on the balanced Figure of Merit (FoM), in which the number of computational steps and the number of required memristors have equal weight, the first and the second proposed multipliers achieve up to 4.6× and 14.9× improvements, respectively, compared to the existing designs in 64-bit multiplication.
采用SIXOR和TMSL逻辑的高速低成本阵列记忆乘法器
记忆系统有许多有前途的特性,使它们既适用于存储也适用于计算。忆阻器可以进行逻辑运算,它们可以用作数字电路中的基本结构,如加法器和乘法器。本文首先利用单周期忆阻器异或(SIXOR)和三忆阻器状态逻辑(TMSL)门,利用这两种逻辑的优点,提出了一种新的快速低成本全加法器(FA)。然后,将所提出的FA作为两个新的阵列乘法器的基本单元之一。第一个提出的乘法器是这样设计的,在现有的设计中,它具有最低的计算步骤(延迟)。与现有设计相比,该设计的延迟平均降低了70%左右。第二种提议的乘法器,作为低成本设计,由于更有效地重用现有资源,需要非常少的忆阻器数量,同时仍然具有低延迟。与最先进的乘法器相比,该乘法器实现了平均约36%的忆阻降低。基于分析,两种提出的阵列乘法器都具有显著的效率优势,与基于不同性能因数(FoMs)的最新设计相比。例如,基于平衡的优点图(FoM),其中计算步数和所需忆阻器的数量具有相同的权重,与现有64位乘法设计相比,第一种和第二种提出的乘法器分别实现了4.6倍和14.9倍的改进。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
IEEE Transactions on Nanotechnology
IEEE Transactions on Nanotechnology 工程技术-材料科学:综合
CiteScore
4.80
自引率
8.30%
发文量
74
审稿时长
8.3 months
期刊介绍: The IEEE Transactions on Nanotechnology is devoted to the publication of manuscripts of archival value in the general area of nanotechnology, which is rapidly emerging as one of the fastest growing and most promising new technological developments for the next generation and beyond.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信
小红书