Article(id=1327012801966617510, tenantId=1146029695717560320, journalId=1324759914146529478, issueId=1327012646391493336, articleNumber=null, orderNo=null, doi=10.7544/issn1000-1239.202550116, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1740499200000, receivedDateStr=2025-02-26, revisedDate=1744646400000, revisedDateStr=2025-04-15, acceptedDate=null, acceptedDateStr=null, onlineDate=1794314106693, onlineDateStr=2026-11-10, pubDate=1748793600000, pubDateStr=2025-06-02, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1794314106693, onlineIssueDateStr=2026-11-10, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1794314106693, creator=13701087609, updateTime=1794314106693, updator=13701087609, issue=Issue{id=1327012646391493336, tenantId=1146029695717560320, journalId=1324759914146529478, year='2025', volume='62', issue='6', pageStart='1331', pageEnd='1593', issueExtLink='null', onlineDate='null', pubDate='1748793600000', pubDateStr='2025-06-02', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1794314069601, creator='13701087609', updateTime=1794376817187, updator='13701087609', preIssue=null, nextIssue=null, articleTotal=null, ext={EN=IssueExt(id=1327275829270262152, tenantId=1146029695717560320, journalId=1324759914146529478, issueId=1327012646391493336, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1327275829270262153, tenantId=1146029695717560320, journalId=1324759914146529478, issueId=1327012646391493336, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null, downloadFileDto=null}, startPage=1581, endPage=1593, ext={EN=ArticleExt(id=1327012802528654247, articleId=1327012801966617510, tenantId=1146029695717560320, journalId=1324759914146529478, language=EN, title=A Reconfigurable Single-Precision Approximate Floating-Point Multiplier Design, columnId=1327012676955386745, journalTitle=Journal of Computer Research and Development, columnName=Chip Architecture Design, runingTitle=null, highlight=null, articleAbstract=

The rapid development of artificial intelligence (AI) and Internet of things (IoT) technologies has increased demands for the computing energy efficiency, and terminal devices are also facing great challenges in terms of hardware resource overhead. In order to solve the problem of energy efficiency, new designs of approximate computing units with low power have been widely studied. In application scenarios such as digital signal processing and image processing, there are a large number of floating-point operations. These applications consume a lot of hardware resources. However, they have certain fault tolerance characteristics and do not have to perform accurate calculations. In this paper, we propose a 32-bit single-precision reconfigurable approximate floating-point multiplier design method based on multiplication to shift addition (MTA) and asymmetric truncation. Firstly, we utilize an efficient and low-power approximate algorithm that transforms multiplication into shift-addition for partial operands. Secondly, to ensure a balance between accuracy and cost, we propose asymmetric truncation processing for the most significant bits (MSBs) and perform precise calculations on the truncated retained parts. The design achieves high precision and low power. A wide design space is generated by combining the MTA algorithm with different bit widths and varying the number of rows in the partial product array after truncation. This allows for various trade-offs between accuracy and cost. Compared with the exact floating-point multiplier, MTA5T5 has an accuracy loss of mean relative error distance (MRED) of only about 0.32%, a power reduction of 85.80%, and an area reduction of about 79.53%. The accuracy loss of MTA3T3 with lower accuracy is about 1.92%, while the power and area decrease to 90.55% and 85.80%, respectively. Ultimately, we test the application of FIR filtering and image processing, and the results show that the proposed design has significant advantages in terms of accuracy.

, authors=Pengcheng Li1, 2, Libo Huang1, 2, *, Gang Chen3, Mingche Lai1, 2, Lin Deng1, 2, Wei Liu1, 2, Qianming Yang1, 2, Yongwen Wang1, 2, authorsList=Pengcheng Li, Libo Huang, Gang Chen, Mingche Lai, Lin Deng, Wei Liu, Qianming Yang, Yongwen Wang, authorCompany=null, correspAuthors=Libo Huang, authorNote=

born in 1995. Master candidate. His main research interest includes computer architecture

born in 1983. PhD, professor. Senior member of CCF. His main research interests include computer architecture and microprocessor design

born in 1985. PhD, professor. Senior member of CCF. His main research interests include artificial intelligence, accelerator design, and embedded systems

born in 1982. PhD, professor. Member of CCF. His main research interest includes computer architecture

born in 1980. PhD, associate professor. Member of CCF. His main research interests include computer architecture and high-performance processor design

born in 1986. PhD, assistant professor. Member of CCF. His main research interest includes microprocessor design

born in 1984. PhD, assistant professor. Member of CCF. His main research interest includes computer architecture

born in 1977. PhD, professor. Senior member of CCF. His main research interest includes microprocessor architecture

, correspAuthorsNote=null, copyrightStatement=Copyright © 2025 Journal of Computer Research and Development. All rights reserved., copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, fund=null), CN=ArticleExt(id=1327012812712424409, articleId=1327012801966617510, tenantId=1146029695717560320, journalId=1324759914146529478, language=CN, title=一种可重构的单精度近似浮点乘法器设计, columnId=1327012677139936123, journalTitle=计算机研究与发展, columnName=芯片架构设计, runingTitle=null, highlight=null, articleAbstract=

人工智能(AI)和物联网(IoT)技术的迅速发展,对计算能效提出了更高的要求,终端设备在硬件资源开销方面同样面临巨大挑战. 为了应对能效问题,新型低功耗近似计算单元的设计得到了广泛研究. 在数字信号处理和图像处理等应用场景中,存在大量的浮点运算. 这些应用消耗了大量的硬件资源,但它们具有一定的容错性,没有必要进行完全精确的计算. 据此,提出了一种基于移位近似算法MTA(multiplication to shift addition)和非对称截断的单精度可重构近似浮点乘法器设计方法. 首先,采用了一种低功耗的近似算法MTA,将部分操作数的乘法运算转换为移位加法. 其次,为了在精度和成本之间取得平衡,设计了针对操作数高有效位的非对称截断处理,并对截断后保留的部分进行精确计算. 通过采用不同位宽的MTA近似计算和改变截断后部分积阵列的行数,生成了广阔的设计空间,从而可以在精度和成本之间进行多种权衡调整. 与精确浮点乘法器相比,所提出设计MTA5T5的精度损失(MRED)仅约为0.32%,功耗降低了85.80%,面积减少了79.53%. 对于精度较低的MTA3T3,其精度损失约为1.92%,而功耗和面积分别降低了90.55%和85.80%. 最后,进行了FIR滤波和图像处理的应用测试,结果表明所提出的设计在精度和开销方面具有显著优势.

, authors=李鹏程1, 2, 黄立波1, 2, *, 陈刚3, 赖明澈1, 2, 邓林1, 2, 刘威1, 2, 杨乾明1, 2, 王永文1, 2, authorsList=李鹏程, 黄立波, 陈刚, 赖明澈, 邓林, 刘威, 杨乾明, 王永文, authorCompany=null, correspAuthors=黄立波, authorNote=

1995年生. 硕士研究生. 主要研究方向为计算机体系结构

1983年生. 博士,研究员. CCF高级会员. 主要研究方向为计算机体系结构、微处理器设计

1985年生. 博士,教授. CCF高级会员. 主要研究领域为人工智能、加速器设计、嵌入式系统

1982年生. 博士,教授. CCF会员. 主要研究方向为计算机体系结构

1980年生. 博士,副研究员. CCF会员. 主要研究方向为计算机体系结构和高性能处理器设计

1986年生. 博士,助理研究员. CCF会员. 主要研究方向为微处理器设计

1984年生. 博士,助理研究员. CCF会员. 主要研究方向为计算机体系结构

1977年生. 博士,研究员. CCF高级会员. 主要研究方向为微处理器体系结构

, correspAuthorsNote=
黄立波()
, copyrightStatement=版权所有 © 2025 计算机研究与发展编辑部, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=yi58Yculh1NoICes8gZ9EQ==, magXml=DzsBmjk4VyPz30swDabn6A==, pdfUrl=null, pdf=jSQkzNvNJ95t1ODm/EWOtw==, pdfFileSize=2218985, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=p0AI43CYqJiTNp8ipTDPyQ==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=5qg1buPYQGu/mLEgS4HZCQ==, mapNumber=null, fund=null)}, authors=[Author(id=1327012814981542884, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=lipengcheng@nudt.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327012815094789096, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012814981542884, language=EN, stringName=Pengcheng Li, firstName=Pengcheng, middleName=null, lastName=Li, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073
2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073, bio={"content":"

born in 1995. Master candidate. His main research interest includes computer architecture

"}, bioImg=null, bioContent=

born in 1995. Master candidate. His main research interest includes computer architecture

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327012815182869481, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012814981542884, language=CN, stringName=李鹏程, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1国防科技大学计算机学院 长沙 410073
2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073, bio={"img":"XkiSHQBjAfGLJh7tRsBcpg==","content":"

1995年生. 硕士研究生. 主要研究方向为计算机体系结构

"}, bioImg=XkiSHQBjAfGLJh7tRsBcpg==, bioContent=

1995年生. 硕士研究生. 主要研究方向为计算机体系结构

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327012813018608602, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=1, ext=[AuthorCompanyExt(id=1327012813031191515, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073), AuthorCompanyExt(id=1327012813043774428, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1国防科技大学计算机学院 长沙 410073)]), AuthorCompany(id=1327012813115077597, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=2, ext=[AuthorCompanyExt(id=1327012813123466206, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073), AuthorCompanyExt(id=1327012813140243423, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073)])]), Author(id=1327012815304504299, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=libohuang@nudt.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1327012815463887854, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012815304504299, language=EN, stringName=Libo Huang, firstName=Libo, middleName=null, lastName=Huang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, *, address=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073
2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073, bio={"content":"

born in 1983. PhD, professor. Senior member of CCF. His main research interests include computer architecture and microprocessor design

"}, bioImg=null, bioContent=

born in 1983. PhD, professor. Senior member of CCF. His main research interests include computer architecture and microprocessor design

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327012815593911279, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012815304504299, language=CN, stringName=黄立波, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, *, address=1国防科技大学计算机学院 长沙 410073
2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073, bio={"img":"oVVn4uYkw0nOEhWEnUSowg==","content":"

1983年生. 博士,研究员. CCF高级会员. 主要研究方向为计算机体系结构、微处理器设计

"}, bioImg=oVVn4uYkw0nOEhWEnUSowg==, bioContent=

1983年生. 博士,研究员. CCF高级会员. 主要研究方向为计算机体系结构、微处理器设计

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327012813018608602, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=1, ext=[AuthorCompanyExt(id=1327012813031191515, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073), AuthorCompanyExt(id=1327012813043774428, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1国防科技大学计算机学院 长沙 410073)]), AuthorCompany(id=1327012813115077597, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=2, ext=[AuthorCompanyExt(id=1327012813123466206, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073), AuthorCompanyExt(id=1327012813140243423, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073)])]), Author(id=1327012815753294834, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327012815849763828, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012815753294834, language=EN, stringName=Gang Chen, firstName=Gang, middleName=null, lastName=Chen, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou 510275, bio={"content":"

born in 1985. PhD, professor. Senior member of CCF. His main research interests include artificial intelligence, accelerator design, and embedded systems

"}, bioImg=null, bioContent=

born in 1985. PhD, professor. Senior member of CCF. His main research interests include artificial intelligence, accelerator design, and embedded systems

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327012815967204341, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012815753294834, language=CN, stringName=陈刚, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3中山大学计算机学院 广州 510275, bio={"img":"hC8QzoS3fvEjtAATmzvv+w==","content":"

1985年生. 博士,教授. CCF高级会员. 主要研究领域为人工智能、加速器设计、嵌入式系统

"}, bioImg=hC8QzoS3fvEjtAATmzvv+w==, bioContent=

1985年生. 博士,教授. CCF高级会员. 主要研究领域为人工智能、加速器设计、嵌入式系统

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327012813232518112, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=3, ext=[AuthorCompanyExt(id=1327012813240906721, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813232518112, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou 510275), AuthorCompanyExt(id=1327012813249295330, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813232518112, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3中山大学计算机学院 广州 510275)])]), Author(id=1327012816046896119, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327012816197891066, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012816046896119, language=EN, stringName=Mingche Lai, firstName=Mingche, middleName=null, lastName=Lai, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073
2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073, bio={"content":"

born in 1982. PhD, professor. Member of CCF. His main research interest includes computer architecture

"}, bioImg=null, bioContent=

born in 1982. PhD, professor. Member of CCF. His main research interest includes computer architecture

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327012816298554363, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012816046896119, language=CN, stringName=赖明澈, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1国防科技大学计算机学院 长沙 410073
2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073, bio={"img":"Q4+WdaJLwsuRe7eBiVoKyw==","content":"

1982年生. 博士,教授. CCF会员. 主要研究方向为计算机体系结构

"}, bioImg=Q4+WdaJLwsuRe7eBiVoKyw==, bioContent=

1982年生. 博士,教授. CCF会员. 主要研究方向为计算机体系结构

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327012813018608602, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=1, ext=[AuthorCompanyExt(id=1327012813031191515, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073), AuthorCompanyExt(id=1327012813043774428, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1国防科技大学计算机学院 长沙 410073)]), AuthorCompany(id=1327012813115077597, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=2, ext=[AuthorCompanyExt(id=1327012813123466206, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073), AuthorCompanyExt(id=1327012813140243423, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073)])]), Author(id=1327012816399217662, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327012816470519808, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012816399217662, language=EN, stringName=Lin Deng, firstName=Lin, middleName=null, lastName=Deng, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073
2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073, bio={"content":"

born in 1980. PhD, associate professor. Member of CCF. His main research interests include computer architecture and high-performance processor design

"}, bioImg=null, bioContent=

born in 1980. PhD, associate professor. Member of CCF. His main research interests include computer architecture and high-performance processor design

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327012816554405889, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012816399217662, language=CN, stringName=邓林, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1国防科技大学计算机学院 长沙 410073
2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073, bio={"img":"aBW6pj+bfnk07FHnV3MwHA==","content":"

1980年生. 博士,副研究员. CCF会员. 主要研究方向为计算机体系结构和高性能处理器设计

"}, bioImg=aBW6pj+bfnk07FHnV3MwHA==, bioContent=

1980年生. 博士,副研究员. CCF会员. 主要研究方向为计算机体系结构和高性能处理器设计

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327012813018608602, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=1, ext=[AuthorCompanyExt(id=1327012813031191515, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073), AuthorCompanyExt(id=1327012813043774428, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1国防科技大学计算机学院 长沙 410073)]), AuthorCompany(id=1327012813115077597, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=2, ext=[AuthorCompanyExt(id=1327012813123466206, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073), AuthorCompanyExt(id=1327012813140243423, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073)])]), Author(id=1327012816642486275, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, orderNo=5, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327012816743149574, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012816642486275, language=EN, stringName=Wei Liu, firstName=Wei, middleName=null, lastName=Liu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073
2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073, bio={"content":"

born in 1986. PhD, assistant professor. Member of CCF. His main research interest includes microprocessor design

"}, bioImg=null, bioContent=

born in 1986. PhD, assistant professor. Member of CCF. His main research interest includes microprocessor design

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327012816831229959, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012816642486275, language=CN, stringName=刘威, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1国防科技大学计算机学院 长沙 410073
2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073, bio={"img":"yXFEyqx5KRH7+Ek6IYhP/g==","content":"

1986年生. 博士,助理研究员. CCF会员. 主要研究方向为微处理器设计

"}, bioImg=yXFEyqx5KRH7+Ek6IYhP/g==, bioContent=

1986年生. 博士,助理研究员. CCF会员. 主要研究方向为微处理器设计

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327012813018608602, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=1, ext=[AuthorCompanyExt(id=1327012813031191515, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073), AuthorCompanyExt(id=1327012813043774428, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1国防科技大学计算机学院 长沙 410073)]), AuthorCompany(id=1327012813115077597, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=2, ext=[AuthorCompanyExt(id=1327012813123466206, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073), AuthorCompanyExt(id=1327012813140243423, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073)])]), Author(id=1327012817040945161, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, orderNo=6, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327012817242271757, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012817040945161, language=EN, stringName=Qianming Yang, firstName=Qianming, middleName=null, lastName=Yang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073
2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073, bio={"content":"

born in 1984. PhD, assistant professor. Member of CCF. His main research interest includes computer architecture

"}, bioImg=null, bioContent=

born in 1984. PhD, assistant professor. Member of CCF. His main research interest includes computer architecture

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327012817426821134, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012817040945161, language=CN, stringName=杨乾明, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1国防科技大学计算机学院 长沙 410073
2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073, bio={"img":"twglYOpZKZmRlaM+ZFkKrQ==","content":"

1984年生. 博士,助理研究员. CCF会员. 主要研究方向为计算机体系结构

"}, bioImg=twglYOpZKZmRlaM+ZFkKrQ==, bioContent=

1984年生. 博士,助理研究员. CCF会员. 主要研究方向为计算机体系结构

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327012813018608602, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=1, ext=[AuthorCompanyExt(id=1327012813031191515, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073), AuthorCompanyExt(id=1327012813043774428, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1国防科技大学计算机学院 长沙 410073)]), AuthorCompany(id=1327012813115077597, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=2, ext=[AuthorCompanyExt(id=1327012813123466206, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073), AuthorCompanyExt(id=1327012813140243423, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073)])]), Author(id=1327012817527484432, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, orderNo=7, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327012817712033811, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012817527484432, language=EN, stringName=Yongwen Wang, firstName=Yongwen, middleName=null, lastName=Wang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073
2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073, bio={"content":"

born in 1977. PhD, professor. Senior member of CCF. His main research interest includes microprocessor architecture

"}, bioImg=null, bioContent=

born in 1977. PhD, professor. Senior member of CCF. His main research interest includes microprocessor architecture

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327012819402338325, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012817527484432, language=CN, stringName=王永文, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1国防科技大学计算机学院 长沙 410073
2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073, bio={"img":"5FWDzfAQ1CAPCC8WGjZzNQ==","content":"

1977年生. 博士,研究员. CCF高级会员. 主要研究方向为微处理器体系结构

"}, bioImg=5FWDzfAQ1CAPCC8WGjZzNQ==, bioContent=

1977年生. 博士,研究员. CCF高级会员. 主要研究方向为微处理器体系结构

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327012813018608602, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=1, ext=[AuthorCompanyExt(id=1327012813031191515, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073), AuthorCompanyExt(id=1327012813043774428, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1国防科技大学计算机学院 长沙 410073)]), AuthorCompany(id=1327012813115077597, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=2, ext=[AuthorCompanyExt(id=1327012813123466206, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073), AuthorCompanyExt(id=1327012813140243423, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073)])])], keywords=[Keyword(id=1327012819578499094, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, orderNo=1, keyword=approximate floating-point multiplier), Keyword(id=1327012819666579480, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, orderNo=2, keyword=MTA), Keyword(id=1327012819746271257, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, orderNo=3, keyword=asymmetric truncation), Keyword(id=1327012819834351642, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, orderNo=4, keyword=accuracy), Keyword(id=1327012819914043419, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, orderNo=5, keyword=cost), Keyword(id=1327012819997929500, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, orderNo=1, keyword=近似浮点乘法器), Keyword(id=1327012820077621277, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, orderNo=2, keyword=MTA), Keyword(id=1327012820161507358, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, orderNo=3, keyword=非对称截断), Keyword(id=1327012820253782047, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, orderNo=4, keyword=精度), Keyword(id=1327012820371222560, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, orderNo=5, keyword=开销)], refs=[Reference(id=1327012826817867853, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=1, rfOrder=0, authorNames=null, journalName=null, refType=null, unstructuredReference=Al-Fuqaha A, Guizani M, Mohammadi M, et al. Internet of things: A survey on enabling technologies, protocols, and applications[J]. IEEE Communications Surveys & Tutorials, 2015, 17(4): 2347−2376, articleTitle=null, refAbstract=null), Reference(id=1327012828499783758, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=2, rfOrder=1, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhuo Cheng, Luo Shaoheng, Gan Houle, et al. Noise-aware DVFS for efficient transitions on battery-powered IoT devices[J]. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2020, 39(7): 1498−1510, articleTitle=null, refAbstract=null), Reference(id=1327012828587864143, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=3, rfOrder=2, authorNames=null, journalName=null, refType=null, unstructuredReference=丁文隆,汪承宁,童薇. 基于自选尾数压缩的高能效浮点忆阻存内处理系统[J]. 计算机研究与发展,2022,59(3):533−552, articleTitle=null, refAbstract=null), Reference(id=1327012828734664784, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=3, rfOrder=3, authorNames=null, journalName=null, refType=null, unstructuredReference=Ding Wenlong, Wang Chengning, Tong Wei. Energy efficient floating-point memristive in-memory processing system based on self-selective mantissa compaction[J]. Journal of Computer Research and Development, 2022, 59(3): 533−552 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327012828877271121, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=4, rfOrder=4, authorNames=null, journalName=null, refType=null, unstructuredReference=Liu Weiqiang, Lombardi F, Shulte M. A retrospective and prospective view of approximate computing[J]. Proceedings of the IEEE, 2020, 108(3): 394−399, articleTitle=null, refAbstract=null), Reference(id=1327012828952768594, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=5, rfOrder=5, authorNames=null, journalName=null, refType=null, unstructuredReference=Lundstrom M. Moore’s law forever?[J]. Science, 2003, 299(5604): 210−211, articleTitle=null, refAbstract=null), Reference(id=1327012829095374931, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=6, rfOrder=6, authorNames=null, journalName=null, refType=null, unstructuredReference=Shafique M, Hafiz R, Rehman S, et al. Invited: Cross-layer approximate computing: From logic to architectures[C]//Proc of the 53rd Annual Design Automation Conf. New York: ACM, 2016: 1−6, articleTitle=null, refAbstract=null), Reference(id=1327012829263147092, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=7, rfOrder=7, authorNames=null, journalName=null, refType=null, unstructuredReference=Han Jie, Orshansky M. Approximate computing: An emerging paradigm for energy-efficient design[C]//Proc of IEEE European Test Symp (ETS). Piscataway, NJ: IEEE, 2013: 1−6, articleTitle=null, refAbstract=null), Reference(id=1327012829342838869, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=8, rfOrder=8, authorNames=null, journalName=null, refType=null, unstructuredReference=Deng Jianing, Shi Zhiguo, Zhuo Cheng. Energy-efficient real-time UAV object detection on embedded platforms[J]. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2020, 39(10): 3123−3127, articleTitle=null, refAbstract=null), Reference(id=1327012829477056598, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=9, rfOrder=9, authorNames=null, journalName=null, refType=null, unstructuredReference=Gupta V, Mohapatra D, Park S P, et al. IMPACT: Imprecise adders for low-power approximate computing[C]//Proc of Int Symp on Low Power Electronics and Design. Piscataway, NJ: IEEE, 2011: 409−414, articleTitle=null, refAbstract=null), Reference(id=1327012829602885719, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=10, rfOrder=10, authorNames=null, journalName=null, refType=null, unstructuredReference=Venkataramani S, Chippa V K, Chakradhar S T, et al. Quality programmable vector processors for approximate computing[C]//Proc of the 46th Annual IEEE/ACM Int Symp on Microarchitecture. New York: ACM, 2013: 1−12, articleTitle=null, refAbstract=null), Reference(id=1327012829787435096, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=11, rfOrder=11, authorNames=null, journalName=null, refType=null, unstructuredReference=Liang Jian, Tessier R, Mencer O. Floating point unit generation and evaluation for FPGAs[C]//Proc of Annual IEEE Symp on Field-Programmable Custom Computing Machines. Piscataway, NJ: IEEE, 2003: 185−194, articleTitle=null, refAbstract=null), Reference(id=1327012829930041434, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=12, rfOrder=12, authorNames=null, journalName=null, refType=null, unstructuredReference=Imani M, Masich M, Peroni D, et al. CANNA: Neural network acceleration using configurable approximation on GPGPU[C]//Proc of the 23rd Asia and South Pacific Design Automation Conf (ASP-DAC). Piscataway, NJ: IEEE, 2018: 682−689, articleTitle=null, refAbstract=null), Reference(id=1327012830013927515, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=13, rfOrder=13, authorNames=null, journalName=null, refType=null, unstructuredReference=Imani M, Garcia R, Gupta S, et al. RMAC: Runtime configurable floating point multiplier for approximate computing[C]//Proc of the Int Symp on Low Power Electronics and Design. New York: ACM, 2018: 1−6, articleTitle=null, refAbstract=null), Reference(id=1327012830106202204, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=14, rfOrder=14, authorNames=null, journalName=null, refType=null, unstructuredReference=Zuras D , Cowlishaw M , Aiken A , et al. IEEE Standard for Floating-Point Arithmetic[J]. IEEE Std 754−2008, 2008: 1−70, articleTitle=null, refAbstract=null), Reference(id=1327012830219448413, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=15, rfOrder=15, authorNames=null, journalName=null, refType=null, unstructuredReference=Lin C H, Lin I C. High accuracy approximate multiplier with error correction[C]//Proc of Int Conf on Computer Design (ICCD). Piscataway, NJ: IEEE, 2013: 33−38, articleTitle=null, refAbstract=null), Reference(id=1327012830311723102, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=16, rfOrder=16, authorNames=null, journalName=null, refType=null, unstructuredReference=Bhardwaj K, Mane P S, Henkel J. Power- and area-efficient approximate wallace tree multiplier for error- resilient systems[C]//Proc of 15th Int Symp on Quality Electronic Design. Piscataway, NJ: IEEE, 2014: 263−269, articleTitle=null, refAbstract=null), Reference(id=1327012830383026271, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=17, rfOrder=17, authorNames=null, journalName=null, refType=null, unstructuredReference=Liu Cong, Han Jie, Lombardi F. A low-power, high performance approximate multiplier with configurable partial error recovery[C]//Proc of Design, Automation & Test in Europe Conf & Exhibition (DATE). Piscataway, NJ: IEEE, 2014: 1−4, articleTitle=null, refAbstract=null), Reference(id=1327012830479495264, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=18, rfOrder=18, authorNames=null, journalName=null, refType=null, unstructuredReference=Narayanamoorthy S, Moghaddam H A, Liu Z, et al. Energy-efficient approximate multiplication for digital signal processing and classification applications[J]. IEEE Transactions on Very Large Scale Integration Systems, 2015, 23(6): 1180−1184, articleTitle=null, refAbstract=null), Reference(id=1327012830550798433, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=19, rfOrder=19, authorNames=null, journalName=null, refType=null, unstructuredReference=Li Zhenhao, Lu Zhaojun, Jia Wei, et al. Efficient approximate floating-point multiplier with runtime reconfigurable frequency and precision[J]. IEEE Transactions on Circuits and Systems II: Express Briefs, 2024, 71(7): 3533−3537, articleTitle=null, refAbstract=null), Reference(id=1327012830630490210, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=20, rfOrder=20, authorNames=null, journalName=null, refType=null, unstructuredReference=Yao Shangshang, Zhang Liang, Wang Qiong, et al. An efficient hybrid parallel compression approximate multiplier[C]//Proc of Int Conf on Computer Design (ICCD). Piscataway, NJ: IEEE, 2021: 109−116, articleTitle=null, refAbstract=null), Reference(id=1327012830743736419, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=21, rfOrder=21, authorNames=null, journalName=null, refType=null, unstructuredReference=Kumar U A, Bikki P, Veeramachaneni S, et al. Power efficient approximate multiplier architectures for error resilient applications[C]//Proc of IEEE 19th India Council Int Conf (INDICON). Piscataway, NJ: IEEE, 2022: 1−5, articleTitle=null, refAbstract=null), Reference(id=1327012830819233892, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=22, rfOrder=22, authorNames=null, journalName=null, refType=null, unstructuredReference=Park G, Kung J, Lee Y. Simplified compressor and encoder designs for low-cost approximate radix-4 booth multiplier[J]. IEEE Transactions on Circuits and Systems II: Express Briefs, 2023, 70(3): 1154−1158, articleTitle=null, refAbstract=null), Reference(id=1327012830886342757, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=23, rfOrder=23, authorNames=null, journalName=null, refType=null, unstructuredReference=Sabetzadeh F, Moaiyeri M H, Ahmadinejad M. An ultra-efficient approximate multiplier with error compensation for error-resilient applications[J]. IEEE Transactions on Circuits and Systems II: Express Briefs, 2023, 70(2): 776−780, articleTitle=null, refAbstract=null), Reference(id=1327012830974423142, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=24, rfOrder=24, authorNames=null, journalName=null, refType=null, unstructuredReference=Haider M H, Zhang Hao, Ko S B. Decoder reduction approximation scheme for booth multipliers[J]. IEEE Transactions on Computers, 2024, 73(3): 735−746, articleTitle=null, refAbstract=null), Reference(id=1327012831083475047, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=25, rfOrder=25, authorNames=null, journalName=null, refType=null, unstructuredReference=Tan Hongbing, Tong Gan, Huang Libo, et al. Multiple- mode-supporting floating-point FMA unit for deep learning processors[J]. IEEE Transactions on Very Large Scale Integration Systems, 2023, 31(2): 253−266, articleTitle=null, refAbstract=null), Reference(id=1327012831158972520, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=26, rfOrder=26, authorNames=null, journalName=null, refType=null, unstructuredReference=Tan Hongbing, Zhang Jing, Huang Libo, et al. A multi-level parallel integer/floating-point arithmetic architecture for deep learning instructions[C]//Proc of Euro-Par 2023: Parallel Processing. Berlin: Springer, 2023: 260−274, articleTitle=null, refAbstract=null), Reference(id=1327012831234469993, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=27, rfOrder=27, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang Hao, Ko S B. Variable-precision approximate floating-point multiplier for efficient deep learning computation[J]. IEEE Transactions on Circuits and Systems II: Express Briefs, 2022, 69(5): 2503−2507, articleTitle=null, refAbstract=null), Reference(id=1327012832962523242, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=28, rfOrder=28, authorNames=null, journalName=null, refType=null, unstructuredReference=Saggese G, Napoli E, Strollo A G M. CFPM: Run-time configurable floating-point multiplier[C]//Proc of the 18th Conf on PhD Research in Microelectronics and Electronics (PRIME). Piscataway, NJ: IEEE, 2023: 173−176, articleTitle=null, refAbstract=null), Reference(id=1327012833058992235, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=29, rfOrder=29, authorNames=null, journalName=null, refType=null, unstructuredReference=Saadat H, Bokhari H, Parameswaran S. Minimally biased multipliers for approximate integer and floating-point multiplication[J]. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2018, 37(11): 2623−2635, articleTitle=null, refAbstract=null), Reference(id=1327012833251930220, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=30, rfOrder=30, authorNames=null, journalName=null, refType=null, unstructuredReference=Imani M, Peroni D, Rosing T. CFPU: Configurable floating point multiplier for energy-efficient computing[C]//Proc of the 54th Annual Design Automation Conf 2017. New York: ACM, 2017: 1−6, articleTitle=null, refAbstract=null), Reference(id=1327012833331621997, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=31, rfOrder=31, authorNames=null, journalName=null, refType=null, unstructuredReference=Jha C K, Walia S, Kanojia G, et al. FPCAM: Floating point configurable approximate multiplier for error resilient applications[C]//Proc of Int Symp on Circuits and Systems (ISCAS). Piscataway, NJ: IEEE, 2021: 1−5, articleTitle=null, refAbstract=null), Reference(id=1327012833398730862, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=32, rfOrder=32, authorNames=null, journalName=null, refType=null, unstructuredReference=Towhidy A, Omidi R, Mohammadi K. On the design of iterative approximate floating-point multipliers[J]. IEEE Transactions on Computers, 2023, 72(6): 1623−1635, articleTitle=null, refAbstract=null), Reference(id=1327012833495199855, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=33, rfOrder=33, authorNames=null, journalName=null, refType=null, unstructuredReference=Liang Jinghang, Han Jie, Lombardi F. New metrics for the reliability of approximate and probabilistic adders[J]. IEEE Transactions on Computers, 2013, 62(9): 1760−1771, articleTitle=null, refAbstract=null)], funds=null, companyList=[AuthorCompany(id=1327012813018608602, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=1, ext=[AuthorCompanyExt(id=1327012813031191515, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073), AuthorCompanyExt(id=1327012813043774428, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1国防科技大学计算机学院 长沙 410073)]), AuthorCompany(id=1327012813115077597, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=2, ext=[AuthorCompanyExt(id=1327012813123466206, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073), AuthorCompanyExt(id=1327012813140243423, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073)]), AuthorCompany(id=1327012813232518112, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=3, ext=[AuthorCompanyExt(id=1327012813240906721, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813232518112, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou 510275), AuthorCompanyExt(id=1327012813249295330, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813232518112, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3中山大学计算机学院 广州 510275)])], figs=[ArticleFig(id=1327012820543189025, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, label=Figure 1, caption=Representation of 32-bit single-precision floating-point numbers, figureFileSmall=pSI279IHohrrDU4j3OkXjw==, figureFileBig=p0AI43CYqJiTNp8ipTDPyQ==, tableContent=null), ArticleFig(id=1327012820627075106, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, label=图1, caption=32位单精度浮点数表示, figureFileSmall=pSI279IHohrrDU4j3OkXjw==, figureFileBig=p0AI43CYqJiTNp8ipTDPyQ==, tableContent=null), ArticleFig(id=1327012821004562467, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, label=Figure 2, caption=Floating-point multiplication operation process, figureFileSmall=igV2EpGSDP+jdLdTj8tYcQ==, figureFileBig=QdkI5dkDx7GPYXuVRPbEiA==, tableContent=null), ArticleFig(id=1327012821084254244, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, label=图2, caption=浮点乘法运算过程, figureFileSmall=igV2EpGSDP+jdLdTj8tYcQ==, figureFileBig=QdkI5dkDx7GPYXuVRPbEiA==, tableContent=null), ArticleFig(id=1327012821226860581, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, label=Figure 3, caption=Block diagram of simplified design for approximate multiplier, figureFileSmall=4iI91XinzczF1A52wewsXA==, figureFileBig=jfXBcxe0exDm4/4cPYkrBw==, tableContent=null), ArticleFig(id=1327012821335912486, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, label=图3, caption=近似乘法器的简要设计框图, figureFileSmall=4iI91XinzczF1A52wewsXA==, figureFileBig=jfXBcxe0exDm4/4cPYkrBw==, tableContent=null), ArticleFig(id=1327012821449158695, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, label=Figure 4, caption=Approximate computing procedure of MTA4, figureFileSmall=XuZjAyvh2ydhVI6UfM+yig==, figureFileBig=7J3THvtADUqpFNn6URh+jw==, tableContent=null), ArticleFig(id=1327012821558210600, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, label=图4, caption=MTA4近似计算过程, figureFileSmall=XuZjAyvh2ydhVI6UfM+yig==, figureFileBig=7J3THvtADUqpFNn6URh+jw==, tableContent=null), ArticleFig(id=1327012821684039721, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, label=Figure 5, caption=Approximate module design of MTA4, figureFileSmall=hl2iBDYLZX1b1/5AJ0NCog==, figureFileBig=rgoKHk7A6zw1QzCdz/V4rQ==, tableContent=null), ArticleFig(id=1327012821784703018, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, label=图5, caption=MTA4近似模块设计, figureFileSmall=hl2iBDYLZX1b1/5AJ0NCog==, figureFileBig=rgoKHk7A6zw1QzCdz/V4rQ==, tableContent=null), ArticleFig(id=1327012821956669483, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, label=Figure 6, caption=Asymmetric truncation process, figureFileSmall=3r8WoHCOHxbN/8znZb9w9Q==, figureFileBig=1N0KZJ+qZL4OgERFG5tpWQ==, tableContent=null), ArticleFig(id=1327012822048944172, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, label=图6, caption=非对称截断处理, figureFileSmall=3r8WoHCOHxbN/8znZb9w9Q==, figureFileBig=1N0KZJ+qZL4OgERFG5tpWQ==, tableContent=null), ArticleFig(id=1327012822170578989, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, label=Figure 7, caption=Leading-one-detection of MTA4, figureFileSmall=IUutXUSstnlvs/C+y2s7KQ==, figureFileBig=6Gis0ooIaIf1hA/xpzowDw==, tableContent=null), ArticleFig(id=1327012822271242286, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, label=图7, caption=MTA4的前导“1”检测, figureFileSmall=IUutXUSstnlvs/C+y2s7KQ==, figureFileBig=6Gis0ooIaIf1hA/xpzowDw==, tableContent=null), ArticleFig(id=1327012824007684145, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, label=Figure 8, caption=Mantissa approximate calculation process of MTA4T5, figureFileSmall=PjzvUx4WXYDfxgUkRtcMVA==, figureFileBig=BwBPHPSlETBoTIID8mRtLA==, tableContent=null), ArticleFig(id=1327012824095764530, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, label=图8, caption=MTA4T5的尾数近似计算过程, figureFileSmall=PjzvUx4WXYDfxgUkRtcMVA==, figureFileBig=BwBPHPSlETBoTIID8mRtLA==, tableContent=null), ArticleFig(id=1327012824179650611, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, label=Figure 9, caption=Relationship diagram of hardware parameters and MRED for different approximate multipliers, figureFileSmall=ySVMxPi5KZhm9y2ETuoyQA==, figureFileBig=xRBZC+jAxKUE2tCvOCiGSQ==, tableContent=null), ArticleFig(id=1327012824297091124, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, label=图9, caption=不同近似乘法器的硬件参数与MRED的关系图, figureFileSmall=ySVMxPi5KZhm9y2ETuoyQA==, figureFileBig=xRBZC+jAxKUE2tCvOCiGSQ==, tableContent=null), ArticleFig(id=1327012824448086069, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, label=Figure 10, caption=Relationship diagram of hardware parameters and SNR for different approximate multipliers, figureFileSmall=XTUDLpDR27S7tGf5KLgMSQ==, figureFileBig=El/j26qNteHZJbou2HNEqw==, tableContent=null), ArticleFig(id=1327012824536166454, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, label=图10, caption=不同近似乘法器的硬件参数与SNR的关系图, figureFileSmall=XTUDLpDR27S7tGf5KLgMSQ==, figureFileBig=El/j26qNteHZJbou2HNEqw==, tableContent=null), ArticleFig(id=1327012824645218359, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, label=Figure 11, caption=Related test images of image self-multiplication, figureFileSmall=ygz0ua+eAqm77XU6sJ8LNg==, figureFileBig=/wPrw4ynTK1twSSrYSzODg==, tableContent=null), ArticleFig(id=1327012824729104440, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, label=图11, caption=图像自乘的相关测试图像, figureFileSmall=ygz0ua+eAqm77XU6sJ8LNg==, figureFileBig=/wPrw4ynTK1twSSrYSzODg==, tableContent=null), ArticleFig(id=1327012824825573434, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, label=Figure 12, caption=Relationship diagram of hardware parameters and PSNR for different approximate multipliers, figureFileSmall=ffo3YjY9oN95Np129glKcw==, figureFileBig=+47aPNSvE8kiqqpkGs9hBA==, tableContent=null), ArticleFig(id=1327012824917848123, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, label=图12, caption=不同近似乘法器的硬件参数与PSNR的关系图, figureFileSmall=ffo3YjY9oN95Np129glKcw==, figureFileBig=+47aPNSvE8kiqqpkGs9hBA==, tableContent=null), ArticleFig(id=1327012825026900028, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, label=Figure 13, caption=Fusion effect of three sets of images, figureFileSmall=FXJ1iukFXBXXgB3enGP4qg==, figureFileBig=JlBdDhaKOHFFlR6gRtBAuw==, tableContent=null), ArticleFig(id=1327012825140146238, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, label=图13, caption=3组图像的融合效果, figureFileSmall=FXJ1iukFXBXXgB3enGP4qg==, figureFileBig=JlBdDhaKOHFFlR6gRtBAuw==, tableContent=null), ArticleFig(id=1327012825219838015, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, label=Figure 14, caption=Relationship diagram of hardware parameters and PSNR for different approximate multipliers, figureFileSmall=ukWP2JigoTdmQjqBwWyurA==, figureFileBig=3m5tLU8E3ZRNBfOGfLrcwQ==, tableContent=null), ArticleFig(id=1327012825324695616, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, label=图14, caption=不同近似乘法器的硬件参数与PSNR的关系图, figureFileSmall=ukWP2JigoTdmQjqBwWyurA==, figureFileBig=3m5tLU8E3ZRNBfOGfLrcwQ==, tableContent=null), ArticleFig(id=1327012825484079169, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, label=Table 1, caption=Error Comparison of Approximate Multipliers, figureFileSmall=null, figureFileBig=null, tableContent=
乘法器MREDNMED乘法器MREDNMED
MTA3T319.175×10−366.69×10−4MTA7T52.350×10−37.73×10−4
MTA4T311.429×10−338.64×10−4HPC[19]3.972×10−313.01×10−4
MTA5T37.767×10−326.24×10−4LPC[19]17.042×10−350.32×10−4
MTA3T48.776×10−329.36×10−4CFPM[28]7.898×10−325.29×10−4
MTA4T45.700×10−318.60×10−4MBM[29]30.720×10−380.58×10−4
MTA5T43.609×10−311.47×10−4CFPU[30]15.103×10−345.62×10−4
MTA3T55.607×10−318.81×10−4PB4W16[32]33.558×10−385.16×10−4
MTA4T53.865×10−312.92×10−4IPB4W16[32]20.004×10−352.11×10−4
MTA5T53.187×10−310.56×10−4PB8W16[32]10.501×10−336.05×10−4
MTA6T52.582×10−38.53×10−4IPB8W16[32]8.200×10−330.65×10−4
), ArticleFig(id=1327012825605713986, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, label=表1, caption=

近似乘法器误差对比

, figureFileSmall=null, figureFileBig=null, tableContent=
乘法器MREDNMED乘法器MREDNMED
MTA3T319.175×10−366.69×10−4MTA7T52.350×10−37.73×10−4
MTA4T311.429×10−338.64×10−4HPC[19]3.972×10−313.01×10−4
MTA5T37.767×10−326.24×10−4LPC[19]17.042×10−350.32×10−4
MTA3T48.776×10−329.36×10−4CFPM[28]7.898×10−325.29×10−4
MTA4T45.700×10−318.60×10−4MBM[29]30.720×10−380.58×10−4
MTA5T43.609×10−311.47×10−4CFPU[30]15.103×10−345.62×10−4
MTA3T55.607×10−318.81×10−4PB4W16[32]33.558×10−385.16×10−4
MTA4T53.865×10−312.92×10−4IPB4W16[32]20.004×10−352.11×10−4
MTA5T53.187×10−310.56×10−4PB8W16[32]10.501×10−336.05×10−4
MTA6T52.582×10−38.53×10−4IPB8W16[32]8.200×10−330.65×10−4
), ArticleFig(id=1327012825714765891, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, label=Table 2, caption=Main Module Overhead of Approximate Multipliers, figureFileSmall=null, figureFileBig=null, tableContent=
模块Power(μW)Area(μm2)
EX-FP32MTA4T5比率EX-FP32MTA4T5比率
注:“n/a”表示不适用对比,“↓”表示相比EX-FP32对应模块下降的百分比.
MTA4n/a22.2n/an/a25.9n/a
前导检测n/a10.6n/an/a76.4n/a
尾数乘法311127891.06%↓4509.6592.286.87%↓
规格化18079.455.89%↓278206.525.72%↓
精确T5n/a72.5n/an/a176.1n/a
), ArticleFig(id=1327012825794457668, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, label=表2, caption=

近似乘法器主要模块开销

, figureFileSmall=null, figureFileBig=null, tableContent=
模块Power(μW)Area(μm2)
EX-FP32MTA4T5比率EX-FP32MTA4T5比率
注:“n/a”表示不适用对比,“↓”表示相比EX-FP32对应模块下降的百分比.
MTA4n/a22.2n/an/a25.9n/a
前导检测n/a10.6n/an/a76.4n/a
尾数乘法311127891.06%↓4509.6592.286.87%↓
规格化18079.455.89%↓278206.525.72%↓
精确T5n/a72.5n/an/a176.1n/a
), ArticleFig(id=1327012825874149445, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, label=Table 3, caption=Comparison of Hardware Parameters of Approximate Multipliers, figureFileSmall=null, figureFileBig=null, tableContent=
乘法器Power/mWArea/μm2Delay/nsPDP(pJ)ADP/(ns×μm2)乘法器Power/mWArea/μm2Delay/nsPDP/pJADP/(ns×μm2)
EX-FP323.46954968.434.7716.549523699.4MTA7T50.57801129.134.732.73395340.8
MTA3T30.3277705.434.701.54023315.5HPC[19]0.61791172.474.732.92275545.8
MTA4T30.3853879.984.741.82634171.1LPC[19]0.4569891.914.752.17034236.6
MTA5T30.4307970.874.732.0374592.2CFPM[28]0.5026981.294.722.37234631.7
MTA3T40.4039867.054.751.91854118.5MBM[29]0.40851098.334.731.93225195.1
MTA4T40.4230906.704.721.99664279.6CFPU[29]2.40285166.013.327.977317151.2
MTA5T40.4593962.474.722.16794542.9PB4W16[32]0.42051135.684.731.98905371.8
MTA3T50.4308928.374.762.05064419.0IPB4W16[32]0.80371846.494.743.80958752.4
MTA4T50.4535948.194.722.14054475.5PB8W16[32]0.63361360.804.732.99696436.6
MTA5T50.49271016.904.732.33054809.9IPB8W16[32]1.14192730.174.745.412612941.0
MTA6T50.53741077.894.732.54195098.4
), ArticleFig(id=1327012825958035526, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, label=表3, caption=

近似乘法器硬件参数对比

, figureFileSmall=null, figureFileBig=null, tableContent=
乘法器Power/mWArea/μm2Delay/nsPDP(pJ)ADP/(ns×μm2)乘法器Power/mWArea/μm2Delay/nsPDP/pJADP/(ns×μm2)
EX-FP323.46954968.434.7716.549523699.4MTA7T50.57801129.134.732.73395340.8
MTA3T30.3277705.434.701.54023315.5HPC[19]0.61791172.474.732.92275545.8
MTA4T30.3853879.984.741.82634171.1LPC[19]0.4569891.914.752.17034236.6
MTA5T30.4307970.874.732.0374592.2CFPM[28]0.5026981.294.722.37234631.7
MTA3T40.4039867.054.751.91854118.5MBM[29]0.40851098.334.731.93225195.1
MTA4T40.4230906.704.721.99664279.6CFPU[29]2.40285166.013.327.977317151.2
MTA5T40.4593962.474.722.16794542.9PB4W16[32]0.42051135.684.731.98905371.8
MTA3T50.4308928.374.762.05064419.0IPB4W16[32]0.80371846.494.743.80958752.4
MTA4T50.4535948.194.722.14054475.5PB8W16[32]0.63361360.804.732.99696436.6
MTA5T50.49271016.904.732.33054809.9IPB8W16[32]1.14192730.174.745.412612941.0
MTA6T50.53741077.894.732.54195098.4
), ArticleFig(id=1327012826041921607, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, label=Table 4, caption=SNR and MSE of FIR Filtering, figureFileSmall=null, figureFileBig=null, tableContent=
乘法器SNR/dBMSE乘法器SNR/dBMSE
MTA3T333.41208.33×10−6MTA7T549.744.86×10−6
MTA4T340.3941.75×10−6HPC[19]46.1311.15×10−6
MTA5T341.5232.19×10−6LPC[19]35.58126.49×10−6
MTA3T441.7730.42×10−6CFPM[28]41.4033.13×10−6
MTA4T444.9614.60×10−6MBM[29]30.17398.62×10−6
MTA5T447.508.14×10−6CFPU[30]37.2487.72×10−6
MTA3T542.3226.77×10−6PB4W16[32]37.5879.78×10−6
MTA4T546.4610.33×10−6IPB4W16[32]36.8694.16×10−6
MTA5T547.398.33×10−6PB8W16[32]42.2426.85×10−6
MTA6T548.766.08×10−6IPB8W16[32]43.8318.94×10−6
), ArticleFig(id=1327012826134196296, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, label=表4, caption=

FIR滤波的SNR和MSE

, figureFileSmall=null, figureFileBig=null, tableContent=
乘法器SNR/dBMSE乘法器SNR/dBMSE
MTA3T333.41208.33×10−6MTA7T549.744.86×10−6
MTA4T340.3941.75×10−6HPC[19]46.1311.15×10−6
MTA5T341.5232.19×10−6LPC[19]35.58126.49×10−6
MTA3T441.7730.42×10−6CFPM[28]41.4033.13×10−6
MTA4T444.9614.60×10−6MBM[29]30.17398.62×10−6
MTA5T447.508.14×10−6CFPU[30]37.2487.72×10−6
MTA3T542.3226.77×10−6PB4W16[32]37.5879.78×10−6
MTA4T546.4610.33×10−6IPB4W16[32]36.8694.16×10−6
MTA5T547.398.33×10−6PB8W16[32]42.2426.85×10−6
MTA6T548.766.08×10−6IPB8W16[32]43.8318.94×10−6
), ArticleFig(id=1327012826239053897, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, label=Table 5, caption=PSNR and SSIM of Image Self-Multiplication, figureFileSmall=null, figureFileBig=null, tableContent=
乘法器AirplaneBaboonBarbaraFemalePeppers平均值
PSNR/dBSSIM/%PSNR/dBSSIM/%PSNR/dBSSIM/%PSNR/dBSSIM/%PSNR/dBSSIM/%PSNR/dBSSIM/%
MTA3T337.127899.8941.309899.9342.141199.9240.370299.8941.134799.9240.416799.91
MTA4T341.233099.9645.194799.9745.865799.9744.659899.9645.019099.9744.394499.97
MTA5T344.286599.9848.097399.9948.700199.9847.194499.9747.971399.9847.249999.98
MTA3T443.104399.9747.279999.9848.236899.9848.027899.9747.276599.9846.785199.98
MTA4T445.705799.9849.989899.9950.922599.9951.358299.9950.069899.9949.609299.99
MTA5T449.731499.9952.844599.9953.6642100.0052.636099.9953.036999.9952.382699.99
MTA3T547.016699.9950.569299.9951.498099.9951.446299.9950.710099.9950.248099.99
MTA4T549.429299.9952.784299.9953.7498100.0052.671599.9952.981199.9952.323299.99
MTA5T550.815999.9954.3882100.0055.0642100.0053.297599.9954.4563100.0053.6044100.00
MTA6T552.4408100.0054.7154100.0055.4167100.0053.361599.9954.9057100.0054.1680100.00
MTA7T552.5110100.0054.9632100.0055.6359100.0053.792999.9955.0829100.0054.3972100.00
HPC[19]49.054999.9953.2100100.0054.5204100.0052.904399.9954.0990100.0052.7577100.00
LPC[19]35.041899.8140.184899.9142.137399.9243.312899.9041.214399.9240.378299.89
CFPM[28]45.323199.9848.967599.9950.460199.9950.204599.9849.247099.9948.840499.99
MBM[29]30.720399.4736.175899.7035.523099.5940.115299.8033.434099.5135.193299.61
CFPU[30]37.247899.9141.913299.9343.423599.9443.777599.9642.526999.9441.777899.94
PB4W16[32]29.038799.4234.583099.6634.561099.5339.463099.7732.506999.4034.030599.56
IPB4W16[32]34.323999.6437.654699.8338.488199.8142.373199.8636.333599.7637.834699.78
PB8W16[32]44.574099.9746.368899.9847.823199.9846.099899.9648.251099.9846.623399.97
IPB8W16[32]45.491699.9848.469799.9949.337799.9846.813399.9750.011999.9948.024899.98
), ArticleFig(id=1327012826381660234, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, label=表5, caption=

图像自乘的PSNR和SSIM

, figureFileSmall=null, figureFileBig=null, tableContent=
乘法器AirplaneBaboonBarbaraFemalePeppers平均值
PSNR/dBSSIM/%PSNR/dBSSIM/%PSNR/dBSSIM/%PSNR/dBSSIM/%PSNR/dBSSIM/%PSNR/dBSSIM/%
MTA3T337.127899.8941.309899.9342.141199.9240.370299.8941.134799.9240.416799.91
MTA4T341.233099.9645.194799.9745.865799.9744.659899.9645.019099.9744.394499.97
MTA5T344.286599.9848.097399.9948.700199.9847.194499.9747.971399.9847.249999.98
MTA3T443.104399.9747.279999.9848.236899.9848.027899.9747.276599.9846.785199.98
MTA4T445.705799.9849.989899.9950.922599.9951.358299.9950.069899.9949.609299.99
MTA5T449.731499.9952.844599.9953.6642100.0052.636099.9953.036999.9952.382699.99
MTA3T547.016699.9950.569299.9951.498099.9951.446299.9950.710099.9950.248099.99
MTA4T549.429299.9952.784299.9953.7498100.0052.671599.9952.981199.9952.323299.99
MTA5T550.815999.9954.3882100.0055.0642100.0053.297599.9954.4563100.0053.6044100.00
MTA6T552.4408100.0054.7154100.0055.4167100.0053.361599.9954.9057100.0054.1680100.00
MTA7T552.5110100.0054.9632100.0055.6359100.0053.792999.9955.0829100.0054.3972100.00
HPC[19]49.054999.9953.2100100.0054.5204100.0052.904399.9954.0990100.0052.7577100.00
LPC[19]35.041899.8140.184899.9142.137399.9243.312899.9041.214399.9240.378299.89
CFPM[28]45.323199.9848.967599.9950.460199.9950.204599.9849.247099.9948.840499.99
MBM[29]30.720399.4736.175899.7035.523099.5940.115299.8033.434099.5135.193299.61
CFPU[30]37.247899.9141.913299.9343.423599.9443.777599.9642.526999.9441.777899.94
PB4W16[32]29.038799.4234.583099.6634.561099.5339.463099.7732.506999.4034.030599.56
IPB4W16[32]34.323999.6437.654699.8338.488199.8142.373199.8636.333599.7637.834699.78
PB8W16[32]44.574099.9746.368899.9847.823199.9846.099899.9648.251099.9846.623399.97
IPB8W16[32]45.491699.9848.469799.9949.337799.9846.813399.9750.011999.9948.024899.98
), ArticleFig(id=1327012826557821003, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=EN, label=Table 6, caption=PSNR and SSIM of Image Fusion, figureFileSmall=null, figureFileBig=null, tableContent=
乘法器Walkbridge×HouseCameraman×LakeAirplane×Clock平均值
PSNR/dBSSIM/%PSNR/dBSSIM/%PSNR/dBSSIM/%PSNR/dBSSIM/%
MTA3T344.762999.9642.846399.9340.051099.9442.553499.94
MTA4T348.518399.9846.809599.9743.877299.9746.401799.97
MTA5T350.878399.9948.655599.9846.358699.9848.630899.98
MTA3T446.548899.9749.596199.9845.532899.9847.225999.98
MTA4T449.820699.9951.997299.9948.185799.9950.001199.99
MTA5T452.926299.9954.627499.9951.420499.9952.991399.99
MTA3T550.880099.9951.718099.9947.246299.9949.948199.99
MTA4T553.460999.9954.217699.9950.568399.9952.748999.99
MTA5T554.8004100.0054.9874100.0052.0498100.0053.9459100.00
MTA6T555.2082100.0055.5066100.0052.6282100.0054.4477100.00
MTA7T555.4174100.0055.7911100.0052.6282100.0054.6122100.00
HPC[19]52.536699.9953.334799.9949.953099.9951.941499.99
LPC[19]40.502199.8943.678799.9436.517899.8840.232999.90
CFPM[28]50.826899.9950.805599.9946.151799.9849.261399.99
MBM[29]38.251699.7134.802699.5032.386199.4935.146899.57
CFPU[30]45.267799.9744.758899.9540.056899.9443.361199.95
PB4W16[32]36.856099.6735.790799.5932.466899.3735.037899.54
IPB4W16[32]35.616499.6335.433799.5732.235099.4934.428499.56
PB8W16[32]50.842799.9949.238399.9845.874099.9748.651799.98
IPB8W16[32]40.006199.9847.694399.9745.432999.9747.377899.97
), ArticleFig(id=1327012826687844428, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, language=CN, label=表6, caption=

图像融合的PSNR和SSIM

, figureFileSmall=null, figureFileBig=null, tableContent=
乘法器Walkbridge×HouseCameraman×LakeAirplane×Clock平均值
PSNR/dBSSIM/%PSNR/dBSSIM/%PSNR/dBSSIM/%PSNR/dBSSIM/%
MTA3T344.762999.9642.846399.9340.051099.9442.553499.94
MTA4T348.518399.9846.809599.9743.877299.9746.401799.97
MTA5T350.878399.9948.655599.9846.358699.9848.630899.98
MTA3T446.548899.9749.596199.9845.532899.9847.225999.98
MTA4T449.820699.9951.997299.9948.185799.9950.001199.99
MTA5T452.926299.9954.627499.9951.420499.9952.991399.99
MTA3T550.880099.9951.718099.9947.246299.9949.948199.99
MTA4T553.460999.9954.217699.9950.568399.9952.748999.99
MTA5T554.8004100.0054.9874100.0052.0498100.0053.9459100.00
MTA6T555.2082100.0055.5066100.0052.6282100.0054.4477100.00
MTA7T555.4174100.0055.7911100.0052.6282100.0054.6122100.00
HPC[19]52.536699.9953.334799.9949.953099.9951.941499.99
LPC[19]40.502199.8943.678799.9436.517899.8840.232999.90
CFPM[28]50.826899.9950.805599.9946.151799.9849.261399.99
MBM[29]38.251699.7134.802699.5032.386199.4935.146899.57
CFPU[30]45.267799.9744.758899.9540.056899.9443.361199.95
PB4W16[32]36.856099.6735.790799.5932.466899.3735.037899.54
IPB4W16[32]35.616499.6335.433799.5732.235099.4934.428499.56
PB8W16[32]50.842799.9949.238399.9845.874099.9748.651799.98
IPB8W16[32]40.006199.9847.694399.9745.432999.9747.377899.97
)], attaches=null, journal=Journal(id=1324759606947315908, delFlag=0, nameCn=计算机研究与发展, nameEn=Journal of Computer Research and Development, nameHistory1=null, nameHistory2=null, issn=1000-1239, eissn=null, cn=11-1777/TP, coden=null, periodic=0, language=CN, oaType=null, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=k2l8zgqwstuxxevFX2qxVA==, journalPrice=null, startedYear=null, abbrevIsoEn=Journal of Computer Research and Development, journalRemark=null, publicationField=null, createdTime=1793776903141, updatedTime=1793850698612, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=J, firstLetterEn=J, subjectCode=Electronic Information, subjectName=null, subjectCodeEn=Electronic Information, subjectNameEn=null, picCn=k2l8zgqwstuxxevFX2qxVA==, picEn=r6k0Ry+RD+c/VXj5QarmDw==, jcr=null, cjcr=null, exts=[JournalExt(id=1325069127710286688, language=CN, name=计算机研究与发展, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1793850698640, updatedTime=1793850698640, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://crad.ict.ac.cn/journalx/authorLogOn.action, submissionEditorUrl=https://crad.ict.ac.cn/journalx/editorLogOn.action, submissionReviewUrl=https://crad.ict.ac.cn/journalx/expertLogOn.action, submissionCeEditorUrl=https://crad.ict.ac.cn/journalx/editorInChiefLogOn.action, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1325069127777395553, language=EN, name=Journal of Computer Research and Development, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1793850698656, updatedTime=1793850698656, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://crad.ict.ac.cn/journalx/authorLogOn.action, submissionEditorUrl=https://crad.ict.ac.cn/journalx/editorLogOn.action, submissionReviewUrl=https://crad.ict.ac.cn/journalx/expertLogOn.action, submissionCeEditorUrl=https://crad.ict.ac.cn/journalx/editorInChiefLogOn.action, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1324759914146529478, websiteList=[Website(id=1325070300139892771, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1324759914146529478, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN, language=CN, createTime=1793850978169, createBy=18614031015, updateTime=1793851040259, updateBy=18614031015, name=计算机研究与发展-中文, tplId=1146099689490845704, title=计算机研究与发展, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1325070893751349360, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=articleTextType, value=kx, createTime=1793851119697, updateTime=1793851119697, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893730377837, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=banner, value=null, createTime=1793851119692, updateTime=1793851119692, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893772320883, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=grayFlag, value=0, createTime=1793851119702, updateTime=1793851119702, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893721989228, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=logo, value=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/file/pic?fileId=U27tDsy717YF7XSPA+Vwdw==, createTime=1793851119690, updateTime=1793851119690, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893784903797, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=minRunFlag, value=0, createTime=1793851119705, updateTime=1793851119705, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893742960751, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/file/pic, createTime=1793851119695, updateTime=1793851119695, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893780709492, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=silenceFlag, value=0, createTime=1793851119704, updateTime=1793851119704, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893738766446, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1793851119694, updateTime=1793851119694, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893755543665, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=themeColor, value=null, createTime=1793851119698, updateTime=1793851119698, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893763932274, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=themeStyle, value=null, createTime=1793851119700, updateTime=1793851119700, creator=18614031015, updator=18614031015)]), Website(id=1325070300580294698, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1324759914146529478, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN, language=EN, createTime=1793850978274, createBy=18614031015, updateTime=1793851036638, updateBy=18614031015, name=计算机研究与发展-英文, tplId=1146101810881728533, title=Journal of Computer Research and Development, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1325070925288321148, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=articleTextType, value=kx, createTime=1793851127216, updateTime=1793851127216, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925263155321, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=banner, value=null, createTime=1793851127210, updateTime=1793851127210, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925309292671, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=grayFlag, value=0, createTime=1793851127221, updateTime=1793851127221, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925254766712, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=logo, value=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/file/pic?fileId=U27tDsy717YF7XSPA+Vwdw==, createTime=1793851127208, updateTime=1793851127208, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925326069889, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=minRunFlag, value=0, createTime=1793851127225, updateTime=1793851127225, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925279932539, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/file/pic, createTime=1793851127214, updateTime=1793851127214, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925317681280, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=silenceFlag, value=0, createTime=1793851127223, updateTime=1793851127223, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925271543930, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1793851127212, updateTime=1793851127212, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925292515453, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=themeColor, value=null, createTime=1793851127217, updateTime=1793851127217, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925300904062, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=themeStyle, value=null, createTime=1793851127219, updateTime=1793851127219, creator=18614031015, updator=18614031015)])], journalTitle=计算机研究与发展, weixinUrl=null, journalUrl=https://crad.ict.ac.cn/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Journal of Computer Research and Development, journalPhotoCn=k2l8zgqwstuxxevFX2qxVA==, journalPhotoEn=r6k0Ry+RD+c/VXj5QarmDw==, journalFirstLetter=J, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=, provinceCode=null, provinceName=null, collectFlag=false, interPubPlatform=, interPubPlatformUrl=null), detailUrlCn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/10.7544/issn1000-1239.202550116, detailUrlEn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/10.7544/issn1000-1239.202550116, pdfUrlCn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/PDF/10.7544/issn1000-1239.202550116, pdfUrlEn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/PDF/10.7544/issn1000-1239.202550116, aliStartDate=0, aliEndDate=0, collectionFlag=false, citedCount=null, citedUrl=null, previewStatus=0, delFlag=0, hasFullText=1, orderTime=1748793600000, fullTextJson=null, articleText=null, reference=null)
收藏切换
一种可重构的单精度近似浮点乘法器设计
李鹏程 1, 2 , 黄立波 1, 2, * , 陈刚 3 , 赖明澈 1, 2 , 邓林 1, 2 , 刘威 1, 2 , 杨乾明 1, 2 , 王永文 1, 2
计算机研究与发展 | 芯片架构设计 2025,62(6): 1581-1593
收起
收藏切换
计算机研究与发展 |芯片架构设计 2025 , 62 (6) : 1581 -1593
一种可重构的单精度近似浮点乘法器设计
全屏
[Author(id=1327012814981542884, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=lipengcheng@nudt.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327012815094789096, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012814981542884, language=EN, stringName=Pengcheng Li, firstName=Pengcheng, middleName=null, lastName=Li, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073
2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073, bio={"content":"

born in 1995. Master candidate. His main research interest includes computer architecture

"}, bioImg=null, bioContent=

born in 1995. Master candidate. His main research interest includes computer architecture

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327012815182869481, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012814981542884, language=CN, stringName=李鹏程, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1国防科技大学计算机学院 长沙 410073
2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073, bio={"img":"XkiSHQBjAfGLJh7tRsBcpg==","content":"

1995年生. 硕士研究生. 主要研究方向为计算机体系结构

"}, bioImg=XkiSHQBjAfGLJh7tRsBcpg==, bioContent=

1995年生. 硕士研究生. 主要研究方向为计算机体系结构

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327012813018608602, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=1, ext=[AuthorCompanyExt(id=1327012813031191515, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073), AuthorCompanyExt(id=1327012813043774428, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1国防科技大学计算机学院 长沙 410073)]), AuthorCompany(id=1327012813115077597, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=2, ext=[AuthorCompanyExt(id=1327012813123466206, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073), AuthorCompanyExt(id=1327012813140243423, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073)])]), Author(id=1327012815304504299, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=libohuang@nudt.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1327012815463887854, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012815304504299, language=EN, stringName=Libo Huang, firstName=Libo, middleName=null, lastName=Huang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, *, address=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073
2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073, bio={"content":"

born in 1983. PhD, professor. Senior member of CCF. His main research interests include computer architecture and microprocessor design

"}, bioImg=null, bioContent=

born in 1983. PhD, professor. Senior member of CCF. His main research interests include computer architecture and microprocessor design

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327012815593911279, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012815304504299, language=CN, stringName=黄立波, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, *, address=1国防科技大学计算机学院 长沙 410073
2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073, bio={"img":"oVVn4uYkw0nOEhWEnUSowg==","content":"

1983年生. 博士,研究员. CCF高级会员. 主要研究方向为计算机体系结构、微处理器设计

"}, bioImg=oVVn4uYkw0nOEhWEnUSowg==, bioContent=

1983年生. 博士,研究员. CCF高级会员. 主要研究方向为计算机体系结构、微处理器设计

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327012813018608602, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=1, ext=[AuthorCompanyExt(id=1327012813031191515, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073), AuthorCompanyExt(id=1327012813043774428, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1国防科技大学计算机学院 长沙 410073)]), AuthorCompany(id=1327012813115077597, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=2, ext=[AuthorCompanyExt(id=1327012813123466206, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073), AuthorCompanyExt(id=1327012813140243423, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073)])]), Author(id=1327012815753294834, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327012815849763828, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012815753294834, language=EN, stringName=Gang Chen, firstName=Gang, middleName=null, lastName=Chen, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou 510275, bio={"content":"

born in 1985. PhD, professor. Senior member of CCF. His main research interests include artificial intelligence, accelerator design, and embedded systems

"}, bioImg=null, bioContent=

born in 1985. PhD, professor. Senior member of CCF. His main research interests include artificial intelligence, accelerator design, and embedded systems

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327012815967204341, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012815753294834, language=CN, stringName=陈刚, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3中山大学计算机学院 广州 510275, bio={"img":"hC8QzoS3fvEjtAATmzvv+w==","content":"

1985年生. 博士,教授. CCF高级会员. 主要研究领域为人工智能、加速器设计、嵌入式系统

"}, bioImg=hC8QzoS3fvEjtAATmzvv+w==, bioContent=

1985年生. 博士,教授. CCF高级会员. 主要研究领域为人工智能、加速器设计、嵌入式系统

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327012813232518112, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=3, ext=[AuthorCompanyExt(id=1327012813240906721, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813232518112, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou 510275), AuthorCompanyExt(id=1327012813249295330, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813232518112, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3中山大学计算机学院 广州 510275)])]), Author(id=1327012816046896119, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327012816197891066, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012816046896119, language=EN, stringName=Mingche Lai, firstName=Mingche, middleName=null, lastName=Lai, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073
2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073, bio={"content":"

born in 1982. PhD, professor. Member of CCF. His main research interest includes computer architecture

"}, bioImg=null, bioContent=

born in 1982. PhD, professor. Member of CCF. His main research interest includes computer architecture

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327012816298554363, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012816046896119, language=CN, stringName=赖明澈, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1国防科技大学计算机学院 长沙 410073
2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073, bio={"img":"Q4+WdaJLwsuRe7eBiVoKyw==","content":"

1982年生. 博士,教授. CCF会员. 主要研究方向为计算机体系结构

"}, bioImg=Q4+WdaJLwsuRe7eBiVoKyw==, bioContent=

1982年生. 博士,教授. CCF会员. 主要研究方向为计算机体系结构

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327012813018608602, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=1, ext=[AuthorCompanyExt(id=1327012813031191515, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073), AuthorCompanyExt(id=1327012813043774428, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1国防科技大学计算机学院 长沙 410073)]), AuthorCompany(id=1327012813115077597, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=2, ext=[AuthorCompanyExt(id=1327012813123466206, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073), AuthorCompanyExt(id=1327012813140243423, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073)])]), Author(id=1327012816399217662, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327012816470519808, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012816399217662, language=EN, stringName=Lin Deng, firstName=Lin, middleName=null, lastName=Deng, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073
2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073, bio={"content":"

born in 1980. PhD, associate professor. Member of CCF. His main research interests include computer architecture and high-performance processor design

"}, bioImg=null, bioContent=

born in 1980. PhD, associate professor. Member of CCF. His main research interests include computer architecture and high-performance processor design

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327012816554405889, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012816399217662, language=CN, stringName=邓林, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1国防科技大学计算机学院 长沙 410073
2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073, bio={"img":"aBW6pj+bfnk07FHnV3MwHA==","content":"

1980年生. 博士,副研究员. CCF会员. 主要研究方向为计算机体系结构和高性能处理器设计

"}, bioImg=aBW6pj+bfnk07FHnV3MwHA==, bioContent=

1980年生. 博士,副研究员. CCF会员. 主要研究方向为计算机体系结构和高性能处理器设计

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327012813018608602, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=1, ext=[AuthorCompanyExt(id=1327012813031191515, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073), AuthorCompanyExt(id=1327012813043774428, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1国防科技大学计算机学院 长沙 410073)]), AuthorCompany(id=1327012813115077597, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=2, ext=[AuthorCompanyExt(id=1327012813123466206, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073), AuthorCompanyExt(id=1327012813140243423, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073)])]), Author(id=1327012816642486275, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, orderNo=5, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327012816743149574, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012816642486275, language=EN, stringName=Wei Liu, firstName=Wei, middleName=null, lastName=Liu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073
2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073, bio={"content":"

born in 1986. PhD, assistant professor. Member of CCF. His main research interest includes microprocessor design

"}, bioImg=null, bioContent=

born in 1986. PhD, assistant professor. Member of CCF. His main research interest includes microprocessor design

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327012816831229959, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012816642486275, language=CN, stringName=刘威, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1国防科技大学计算机学院 长沙 410073
2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073, bio={"img":"yXFEyqx5KRH7+Ek6IYhP/g==","content":"

1986年生. 博士,助理研究员. CCF会员. 主要研究方向为微处理器设计

"}, bioImg=yXFEyqx5KRH7+Ek6IYhP/g==, bioContent=

1986年生. 博士,助理研究员. CCF会员. 主要研究方向为微处理器设计

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327012813018608602, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=1, ext=[AuthorCompanyExt(id=1327012813031191515, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073), AuthorCompanyExt(id=1327012813043774428, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1国防科技大学计算机学院 长沙 410073)]), AuthorCompany(id=1327012813115077597, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=2, ext=[AuthorCompanyExt(id=1327012813123466206, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073), AuthorCompanyExt(id=1327012813140243423, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073)])]), Author(id=1327012817040945161, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, orderNo=6, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327012817242271757, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012817040945161, language=EN, stringName=Qianming Yang, firstName=Qianming, middleName=null, lastName=Yang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073
2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073, bio={"content":"

born in 1984. PhD, assistant professor. Member of CCF. His main research interest includes computer architecture

"}, bioImg=null, bioContent=

born in 1984. PhD, assistant professor. Member of CCF. His main research interest includes computer architecture

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327012817426821134, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012817040945161, language=CN, stringName=杨乾明, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1国防科技大学计算机学院 长沙 410073
2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073, bio={"img":"twglYOpZKZmRlaM+ZFkKrQ==","content":"

1984年生. 博士,助理研究员. CCF会员. 主要研究方向为计算机体系结构

"}, bioImg=twglYOpZKZmRlaM+ZFkKrQ==, bioContent=

1984年生. 博士,助理研究员. CCF会员. 主要研究方向为计算机体系结构

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327012813018608602, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=1, ext=[AuthorCompanyExt(id=1327012813031191515, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073), AuthorCompanyExt(id=1327012813043774428, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1国防科技大学计算机学院 长沙 410073)]), AuthorCompany(id=1327012813115077597, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=2, ext=[AuthorCompanyExt(id=1327012813123466206, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073), AuthorCompanyExt(id=1327012813140243423, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073)])]), Author(id=1327012817527484432, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, orderNo=7, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327012817712033811, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012817527484432, language=EN, stringName=Yongwen Wang, firstName=Yongwen, middleName=null, lastName=Wang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073
2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073, bio={"content":"

born in 1977. PhD, professor. Senior member of CCF. His main research interest includes microprocessor architecture

"}, bioImg=null, bioContent=

born in 1977. PhD, professor. Senior member of CCF. His main research interest includes microprocessor architecture

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327012819402338325, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, authorId=1327012817527484432, language=CN, stringName=王永文, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1国防科技大学计算机学院 长沙 410073
2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073, bio={"img":"5FWDzfAQ1CAPCC8WGjZzNQ==","content":"

1977年生. 博士,研究员. CCF高级会员. 主要研究方向为微处理器体系结构

"}, bioImg=5FWDzfAQ1CAPCC8WGjZzNQ==, bioContent=

1977年生. 博士,研究员. CCF高级会员. 主要研究方向为微处理器体系结构

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327012813018608602, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=1, ext=[AuthorCompanyExt(id=1327012813031191515, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073), AuthorCompanyExt(id=1327012813043774428, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813018608602, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1国防科技大学计算机学院 长沙 410073)]), AuthorCompany(id=1327012813115077597, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, xref=2, ext=[AuthorCompanyExt(id=1327012813123466206, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073), AuthorCompanyExt(id=1327012813140243423, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012801966617510, companyId=1327012813115077597, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073)])])]
李鹏程1, 2 , 黄立波1, 2, * , 陈刚3, 赖明澈1, 2, 邓林1, 2, 刘威1, 2, 杨乾明1, 2, 王永文1, 2
作者信息
  • 1国防科技大学计算机学院 长沙 410073
  • 2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073
  • 3中山大学计算机学院 广州 510275
通讯作者:
黄立波()
作者简介:

1995年生. 硕士研究生. 主要研究方向为计算机体系结构

1983年生. 博士,研究员. CCF高级会员. 主要研究方向为计算机体系结构、微处理器设计

1985年生. 博士,教授. CCF高级会员. 主要研究领域为人工智能、加速器设计、嵌入式系统

1982年生. 博士,教授. CCF会员. 主要研究方向为计算机体系结构

1980年生. 博士,副研究员. CCF会员. 主要研究方向为计算机体系结构和高性能处理器设计

1986年生. 博士,助理研究员. CCF会员. 主要研究方向为微处理器设计

1984年生. 博士,助理研究员. CCF会员. 主要研究方向为计算机体系结构

1977年生. 博士,研究员. CCF高级会员. 主要研究方向为微处理器体系结构

A Reconfigurable Single-Precision Approximate Floating-Point Multiplier Design
Pengcheng Li1, 2 , Libo Huang1, 2, * , Gang Chen3, Mingche Lai1, 2, Lin Deng1, 2, Wei Liu1, 2, Qianming Yang1, 2, Yongwen Wang1, 2
Affiliations
  • 1College of Computer Science and Technology, National University of Defense Technology, Changsha 410073
  • 2Key Laboratory of Advanced Microprocessor Chips and Systems (National University of Defense Technology), Changsha 410073
  • 3School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou 510275
  • born in 1995. Master candidate. His main research interest includes computer architecture

    born in 1983. PhD, professor. Senior member of CCF. His main research interests include computer architecture and microprocessor design

    born in 1985. PhD, professor. Senior member of CCF. His main research interests include artificial intelligence, accelerator design, and embedded systems

    born in 1982. PhD, professor. Member of CCF. His main research interest includes computer architecture

    born in 1980. PhD, associate professor. Member of CCF. His main research interests include computer architecture and high-performance processor design

    born in 1986. PhD, assistant professor. Member of CCF. His main research interest includes microprocessor design

    born in 1984. PhD, assistant professor. Member of CCF. His main research interest includes computer architecture

    born in 1977. PhD, professor. Senior member of CCF. His main research interest includes microprocessor architecture

出版时间: 2025-06-02 doi: 10.7544/issn1000-1239.202550116
文章导航
收藏切换

人工智能(AI)和物联网(IoT)技术的迅速发展,对计算能效提出了更高的要求,终端设备在硬件资源开销方面同样面临巨大挑战. 为了应对能效问题,新型低功耗近似计算单元的设计得到了广泛研究. 在数字信号处理和图像处理等应用场景中,存在大量的浮点运算. 这些应用消耗了大量的硬件资源,但它们具有一定的容错性,没有必要进行完全精确的计算. 据此,提出了一种基于移位近似算法MTA(multiplication to shift addition)和非对称截断的单精度可重构近似浮点乘法器设计方法. 首先,采用了一种低功耗的近似算法MTA,将部分操作数的乘法运算转换为移位加法. 其次,为了在精度和成本之间取得平衡,设计了针对操作数高有效位的非对称截断处理,并对截断后保留的部分进行精确计算. 通过采用不同位宽的MTA近似计算和改变截断后部分积阵列的行数,生成了广阔的设计空间,从而可以在精度和成本之间进行多种权衡调整. 与精确浮点乘法器相比,所提出设计MTA5T5的精度损失(MRED)仅约为0.32%,功耗降低了85.80%,面积减少了79.53%. 对于精度较低的MTA3T3,其精度损失约为1.92%,而功耗和面积分别降低了90.55%和85.80%. 最后,进行了FIR滤波和图像处理的应用测试,结果表明所提出的设计在精度和开销方面具有显著优势.

近似浮点乘法器  /  MTA  /  非对称截断  /  精度  /  开销

The rapid development of artificial intelligence (AI) and Internet of things (IoT) technologies has increased demands for the computing energy efficiency, and terminal devices are also facing great challenges in terms of hardware resource overhead. In order to solve the problem of energy efficiency, new designs of approximate computing units with low power have been widely studied. In application scenarios such as digital signal processing and image processing, there are a large number of floating-point operations. These applications consume a lot of hardware resources. However, they have certain fault tolerance characteristics and do not have to perform accurate calculations. In this paper, we propose a 32-bit single-precision reconfigurable approximate floating-point multiplier design method based on multiplication to shift addition (MTA) and asymmetric truncation. Firstly, we utilize an efficient and low-power approximate algorithm that transforms multiplication into shift-addition for partial operands. Secondly, to ensure a balance between accuracy and cost, we propose asymmetric truncation processing for the most significant bits (MSBs) and perform precise calculations on the truncated retained parts. The design achieves high precision and low power. A wide design space is generated by combining the MTA algorithm with different bit widths and varying the number of rows in the partial product array after truncation. This allows for various trade-offs between accuracy and cost. Compared with the exact floating-point multiplier, MTA5T5 has an accuracy loss of mean relative error distance (MRED) of only about 0.32%, a power reduction of 85.80%, and an area reduction of about 79.53%. The accuracy loss of MTA3T3 with lower accuracy is about 1.92%, while the power and area decrease to 90.55% and 85.80%, respectively. Ultimately, we test the application of FIR filtering and image processing, and the results show that the proposed design has significant advantages in terms of accuracy.

approximate floating-point multiplier  /  MTA  /  asymmetric truncation  /  accuracy  /  cost
李鹏程, 黄立波, 陈刚, 赖明澈, 邓林, 刘威, 杨乾明, 王永文. 一种可重构的单精度近似浮点乘法器设计. 计算机研究与发展, 2025 , 62 (6) : 1581 -1593 . DOI: 10.7544/issn1000-1239.202550116
Pengcheng Li, Libo Huang, Gang Chen, Mingche Lai, Lin Deng, Wei Liu, Qianming Yang, Yongwen Wang. A Reconfigurable Single-Precision Approximate Floating-Point Multiplier Design[J]. Journal of Computer Research and Development, 2025 , 62 (6) : 1581 -1593 . DOI: 10.7544/issn1000-1239.202550116
高性能计算、多媒体和人工智能大模型等新一代信息技术的迅速发展,正以颠覆性创新姿态重塑全球科技图景,从推动科学研究跃迁到赋能千行百业智能化转型,这些技术突破不仅释放了巨大的社会生产力,更催生出全新的产业生态. 然而,技术跃升的背后潜藏着严峻的挑战:计算机系统的总能耗正以惊人的速度增长,以处理不断增加的信息量,提高计算能效已成为解决能耗持续增长的关键[1-3]. 然而随着登纳德缩放定律几乎失效[4],摩尔定律逐渐走到尽头[5],晶体管的尺寸正接近其物理极限,导致“暗硅”[6]等许多关键问题产生,通过缩小晶体管尺寸来大幅降低功耗、面积等资源开销的方式已经难以为继. 为了应对日益增长的性能和功耗挑战,研究人员正在积极探索一种新的计算范式——近似计算[7]. 近似计算通过牺牲一定的精确度,降低了电路的逻辑复杂度,简化了硬件结构,从而大幅降低了功耗和面积等开销.
在数字信号处理和图像处理等计算密集型应用中,乘法是最典型且出现频次最高的操作[8],完全精确的乘法计算占用了极大的功耗和延迟. 同时,这类应用具有一定的容错性,因而近似乘法器在近似计算领域得到越来越多研究人员的关注[9-10]. 尽管浮点运算单元在当前系统中成本较高且运行缓慢,但乘法器中仍然存在大量浮点格式设计[11-13].
近似浮点乘法器经过多年来的发展已经取得了阶段性成果,但仍然存在一些亟待解决的问题. 一方面,设计的近似浮点乘法器可重构性不足以及精度和开销固定,难以适用于不同要求的多种应用场景;另一方面,虽然存在一些精度可调节的近似设计,但是难以实现精度的精细调节,往往都是粗粒度的精度和开销控制. 此外,精度调节单元的加入也大大增加了近似电路设计的复杂性,成本迅速增加. 针对近似浮点乘法器的研究需求,本文提出了一种可重构的近似浮点乘法器设计方案. 在设计阶段,策略性地运用多种截断和MTA算法相结合,充分探索整个设计空间. MTA是一种将乘法运算转换为移位加法运算的近似计算方法,它使用前导“1”检测和移位电路来实现运算,但不会显著增加开销. 所采用的截断方法不同于常见的部分积列截断,而是采用部分积列截断和行截断相结合的非对称截断方式,能够与MTA相互配合实现良好的精度和开销平衡.
实验结果表明,与精确乘法器相比,高精度设计MTA5T5的精度损失仅约为0.32%,功耗延迟积(power-delay product,PDP)降低了85.92%,低精度设计MTA3T3的精度损失达1.9%,但PDP降至90.69%. 与当前先进的近似乘法器相比,所提出的设计在精度和资源效率之间取得了更有力的平衡.
本文的主要贡献有3点:
1)实现了一种通过前导“1”检测和移位相加进行近似乘法计算的结构,避免了过多的门级电路,减少了硬件资源需求和延迟;
2)根据尾数乘法的特点,通过对部分积阵列进行不同的截断,并与MTA算法相结合,广泛探索了近似浮点乘法器的设计空间;
3)利用仿真和综合工具验证了可重构近似浮点乘法器在精度和开销上的优势,并进一步通过应用测试验证了所提出设计的良好效果.
浮点乘法运算中的尾数乘法与定点乘法几乎相同,但是浮点数的动态计算和表示范围更大,导致通常会比整数或定点计算消耗更多的资源. 根据浮点数IEEE 754标准[14],单精度浮点数由1位符号位S(sign)、8位指数位E(exponent)和23位尾数位M(mantissa)共计32位组成,如图1所示.
S决定浮点数是正数(S=0)还是负数(S=1);E是2的幂,作用是对浮点数加权. 由于指数可以为正数,也可以为负数,为了处理负指数的情况,实际的指数值需要加上偏置(Bias)值作为保存在指数段中的值,在单精度浮点数中Bias=127. M是二进制小数,取值范围为0~1,加上隐藏位“1”构成完整的尾数部分1.M. 因此,一个规格化的单精度浮点数X可以用式(1)表示:
$ X = {( - 1)^S} \times (1.M) \times {2^{E - 127}}. $
2个单精度浮点数A、B相乘得到C的过程如图2所示,首先2个操作数的符号位通过异或操作得到结果的符号位,尾数部分通过相乘并舍入到1~2的范围内以产生规格化的表示,然后指数部分通过相加后减去偏置值和尾数计算的舍入情况来得到规格化的指数结果,最终将规格化的S、E、M三部分按照IEEE 754的标准拼接得到最终结果.
目前关于近似乘法器的研究可以分为定点和浮点2种设计类型,研究人员通常在乘法运算的部分积生成和累加2个阶段进行近似设计,主要的技术方法[15-19]包括直接简化乘法运算和改变乘法算法等. 在部分积生成阶段可以通过操作数分割、Booth编码、对数转化等引入近似,部分积累加阶段主要通过截断处理、调整精确和近似压缩的比例、设计使用近似单元等实现近似设计.
在定点近似乘法设计中,Yao等人[20]通过设计预处理单元,将部分积阵列通过预处理调整为稀疏部分和稠密部分,在稀疏部分采用或门压缩以节省开销,稠密部分采用精确压缩以保证精度,但该方法针对部分积阵列行数较多的情况容易引起较大误差. Liu等人[17]设计了一种限制进位传播以实现并行计算的近似加法器,利用加法器树进行部分积的累加,使用近似加法器产生的误差向量来补偿输出中的错误,以实现更高的精度. Kumar等人[21]在部分积约简过程中加入误差校正模块,其根据输入模式生成“1”或“0”,将纠错输出作为部分积精确区域第1列的“随身行李”以提高计算精度. 然而,在上述设计中加入额外的纠错模块不可避免地增加了总体成本和电路的复杂性. 为了不使用额外的误差补偿模块,Park等人[22]提出通过在部分积生成和累加2个阶段强制生成彼此方向相反的误差来综合考虑,首先改进Booth编码使其在部分积生成阶段产生更大的单向误差,并降低硬件成本,然后在部分积累加阶段设计误差方向相反的近似压缩器以补偿来自Booth编码器的误差,从而在不使用复杂的误差恢复电路的情况下将内部误差自然平衡为零. Sabetzadeh等人[23]将部分积阵列的一半低有效位作为常数. 这种设计虽然实现了非常高的硬件效率,但也导致了较高的误差,特别是在输入数据较小的情况下. Haider等人[24]提出了一种称为解码器减少近似(decoder reduction approximation, DRA)的新近似方案,该方案使用较少的N/4解码器来代替传统Booth乘法器中的N/2解码器,从而减少了部分积行数的累加. 但由于该方案的技术类似于部分积阵列行的截断,因此在低位宽乘法器中会导致更大的精度损失. 此外,DRA方案的编码过滤逻辑相对复杂,导致资源开销增加.
相比于定点乘法器,浮点乘法由于其动态范围更广,在数字信号处理、图像处理和深度学习等应用场景中有广泛的应用[25-26]. 在浮点乘法器中,由于尾数乘法的功耗、面积等开销占比巨大,且能较易容忍误差,因而有关浮点乘法器的近似设计主要体现在尾数乘法部分. Zhang等人[27]通过设置精度控制单元以计算Bfloat16浮点乘积的指数部分,然后将其转换为posit编码格式,并生成掩码信号来控制尾数乘法器的近似程度. 在尾数乘法的部分积阵列中,如果掩码位为1,则相应位置将使用压缩器累加部分积,否则将禁用该部分积列的累加计算,从而实现部分积阵列的截断处理. Saggese等人[28]通过动态截断部分积阵列,并使用简单的舍入来减少误差,设计了一种精度可调的浮点乘法器. Saadat等人[29]提出了一种误差可配置近似乘法器,通过在对数相加阶段中加入一个固定修正项来消除误差偏差并保持了较低的硬件复杂度. 类似地,Imani等人[30]提出了另一种可配置的低开销近似浮点乘法器,该乘法器使用自适应选择器和调谐电路评估输入操作数,从而动态识别可能导致的误差,如果错误被认为是可接受的,自适应选择器电路更新指数值,并选择一个输入操作数作为输出尾数,否则切换到精确模式进行计算. Jha等人[31]提出了与Imani等人不同的方法,不再需要单独的精确计算单元,只使用移位和加法运算,具有运行时的可配置性,但是该设计本质上是部分积阵列的行累加,精度和开销均无法达到较好的平衡. Towhidy等人[32]提出了基于伪Booth编码的迭代近似浮点乘法器,该乘法器可以通过编码器基数、截断的位宽以及是否迭代调整精度,但随着精度的提高,开销急剧增加.
从上述相关工作中可知,近似乘法器设计的主要技术思想可以总结为2点:一是确保高权重部分进行精确计算,以保证整体设计的准确性;二是尽可能在低权重部分进行近似计算,以降低硬件成本. 目前许多近似设计存在高权重位部分精确计算冗余或不足,以及低权重位近似计算成本过高的问题,难以实现精度与硬件资源之间的良好平衡. 为解决这一问题,本文提出了一种基于MTA和非对称截断的高精度低功耗近似浮点乘法器设计方案. 该方案主要对2个操作数的高有效位部分进行非对称截断,而其他位则采用低开销的MTA近似算法进行计算. 非对称截断解决了高有效位精确计算冗余的问题,而MTA近似算法解决了低有效位计算成本过高的问题. 通过MTA与截断处理相结合以及利用浮点乘法的特点等进行近似设计,在精度和资源开销之间实现了良好的平衡.
截断处理和MTA算法的多种不同组合使得近似浮点乘法器具有较强的可重构性,形成较大的设计空间,所提出近似乘法器的简化框图如图3所示. 由于浮点乘法的特性,我们在指数位采用精确计算,对尾数相乘采用近似计算. 当对2个操作数的尾数进行截断处理和MTA近似计算后,经过位宽扩展和加法器进行最后的规格化. 在2.2节中,我们将详细描述所提出的乘数器设计方法,并探讨其多种可重构设计的不同效果.
MTA近似算法是一种将乘法操作转变为移位加法操作的近似计算方法. 从单精度浮点数的表达式(式(1))可以看出,尾数乘法计算中的最高有效位始终为“1”,根据这一特征,提出MTA近似算法. 假设存在2个最高有效位均为“1”的4位二进制数A和B,分别可以由式(2)和式(3)表示.
$ A = 1xxx = 1 \times {2^3} + m \text{,} $
$ B = 1xxx = 1 \times {2^3} + n . $
这2个二进制数相乘的结果T如式(4)所示.
$ \begin{split}T = & A \times B = (1 \times {2^3} + m) \times (1 \times {2^3} + n)= \\& 1 \times {2^6} + (m + n) \times {2^3} + m \times n. \end{split} $
从式(2)~(4)可以看出,最高有效位为“1”的2个操作数在相乘的过程中可以转变为部分加法操作,因此,根据这一特点提出了将乘法操作近似为加法运算的MTA近似算法. 4位MTA近似算法(MTA4)的具体计算过程如图4所示,A与B相乘转变为A与B相加. 在2个操作数A和B相加的过程中,只有m和n参与求和运算,并得到和的低3位结果sum. 中间结果的最高有效位始终为“1”,并不需要计算得到. 当m与n和的最高位不产生进位时,通过将中间结果左移3位,然后用m或n来填充低3位以获得最终结果. $ A \times B $的最终结果被近似为式(5).
$ {T_{\rm appro}} \approx A \times B \approx 1 \times {2^6} + (m + n) \times {2^3} + m . $
此时,MTA4近似算法的误差值计算如式(6)所示.
$ Err = A \times B - {T_{\rm appro}} = m \times n - m = (m - 1) \times n. $
当$ m + n $的最高有效位产生进位“1”时,$ A \times B $的最终结果被近似为$ (m + n) \times {2^4} $,即图4中的红色虚线部分所示,通过$ m + n $获得中间结果,然后将中间结果向左移4位并用“0000”补充低4位以获得最终结果. 此时,MTA4近似计算的误差如式(7)所示.
$ \begin{split}Err =& A \times B - {T_{\rm appro}} = 1 \times {2^6} + (m + n) \times {2^3} + m \times n - \\& (m + n) \times {2^4} = 1 \times {2^6} - (m + n) \times {2^3} + m \times n .\end{split} $
MTA4的电路设计较为精简,如图5所示,2个3位的操作数m和n输入4位加法器中,由加法器分别输出$ m + n $的进位以及低3位sum,sum通过位拼接形成2个多路选择器MUX的输入,进位信号通过MUX控制选择输出近似结果.
该方法有效地将复杂的乘法运算转化为简单的移位加法运算,降低了资源消耗,适合乘法中低权重部分的近似计算. 当操作数A和B的位宽发生变化时,上述近似算法依然适用. 在本文中,我们主要对3~7位MTA算法进行近似设计,从算法设计以及实际效果来看,3位及以下的MTA近似算法具有较低的准确性,并且需要经历更多的前导“1”检测,延迟有所增加,位宽大于6的MTA近似算法提高精度的幅度较小,但是硬件资源开销增加相对明显,难以在精度和成本之间取得良好的平衡.
部分积阵列的截断处理在近似设计中是一种常用的方法,大多数截断处理采用的是截断部分积阵列的行或者列的方法,但是为了能够与MTA近似算法相结合,实现良好的精度和开销平衡,我们采用了一种能够结合MTA的截断处理方法,选择在部分积阵列中的行和列同时展开截断操作,大幅降低功耗、面积等开销. 除此之外,相比定点乘法场景,非对称截断处理依然更加适合浮点乘法计算,这主要是由浮点乘法的特性决定,浮点尾数的隐藏位“1”导致高位截断误差呈指数放大,而定点乘法误差线性累积.
在非对称截断处理中,我们采用截断操作数的方式来实现截断处理,而非截断部分积阵列的方式,有效降低了部分积计算所带来的资源开销. 如图6所示,X、Y两个操作数相乘,将操作数X截断并保留高10位,将操作数Y截断并保留高5位. 假设X、Y的最高有效位均为1,当与最高有效位相乘时,可以不使用与门直接得到部分积的计算结果(图6中灰色阴影部分). 然后对截断的操作数在部分积生成过程中进一步截断得到T5部分积. 当Y被进一步截断,只保留高3位、高4位时,生成的部分积分别称为T3和T4. 当部分积阵列为T3时,部分积采用全加器并行计算;当部分积阵列为T4时,部分积采用精确4-2压缩器并行计算;当部分积阵列为T5时,阴影上方的4行部分积采用精确4-2压缩器并行计算,然后将结果与阴影部分积相加得到整个截断操作的结果.
通过非对称截断处理,生成的部分积有利于使用精确加法器或者4-2压缩器等标准压缩部件进行并行压缩,提高了计算性能,并大幅降低了功耗、延迟和面积等. 但是,过多的截断操作也带来了严重的精度损失,当配合MTA近似算法时,这种精度损失会得到明显降低,从而提高精度.
在单精度浮点计算中,由于指数位的计算对精度影响过大,一般在尾数计算中引入近似计算. 在23位尾数中,高有效位对精度的影响更重要,当它们改变时会导致精度大幅波动,因此,在设计尾数近似计算时,尽可能保证高权重位执行精确计算,而低权重位则被近似处理.
在X×Y的过程中,首先固定Y的最高有效位数,这里选择5位最高有效位. 操作数X可以从第23位开始逐渐截断,这里选择截断低14位,X和Y被截断后保留的部分执行精确乘法计算,在截断处理后,再运用MTA算法执行近似计算. 如图7所示,采用MTA4近似算法,选择X的高4位,其最高有效位固定为“1”. Y的高5位已经执行精确计算,其余低位(18~3)采用前导“1”检测,当检测到第一个有效位“1”时,同时选择其后的3位,构成最高有效位为“1”的4位操作数. 当18~3位经过前导“1”检测均为0后,由于2~0位在整体所占重要性较低,因此,采用调优补偿的方式,将MTA的结果近似取值为24’b011.
以MTA4T5为例,整体计算过程如图8所示. 首先选择X的23~14位和Y的23~19位执行精确计算,生成精确结果P1(47~37位). 然后,根据MTA4近似算法,选择X的23~20位高有效位,在Y的18~3位利用前导“1”检测选择最高有效位为“1”的4位操作数,并执行MTA4近似计算,得到近似结果P2. 图8中的P1和P2只扩展到47~23位,这是因为单精度浮点数的尾数部分只保留23位,在规格化的过程中会对部分积阵列的低权重位所产生的结果进行舍入,因此为了进一步降低开销,这里不再进行尾数结果低权重位的舍入,而是将计算过程中的22~0位直接截断,虽然是大比特的直接截断,但是在精度损失较低的前提下,大幅降低了硬件成本. 最后将P1和P2相加,得到尾数乘法的最终结果P3.
本文提出的MTA近似算法和非对称截断处理具有良好的可重构性,不同位宽的MTA近似算法和非对称截断处理相结合能形成广泛的设计空间,进而实现不同的精度和硬件资源开销之间的组合,探索不同的可重构设计有助于寻找最佳设计方案,也为不同精度和开销需求的应用提供了可选择性.
在MTA近似算法中,当位宽小于3时,所提供的精度较低,随着位宽的不断增加,精度逐渐提高,但硬件开销也相应增加. 当MTA的位宽大于6时,精度的提高相对缓慢,而功耗却仍持续线性增长. 此外,非对称截断也具备一定的可配置性,本文固定X的高10位,探索了Y的不同截断,也可一定程度上固定Y的位宽,调整X的截断作为变量进行研究,而两者的可操作性和设计形式类似,故不再单独作为空间探索的设计. 因此,在探索设计空间时,本文主要应用和测试了MTA3,MTA4和MTA5三种近似算法,并相应采用了T3,T4和T5三种不同的非对称截断处理,产生了9种不同的近似设计,分别为MTA3T3,MTA3T4,MTA3T5,MTA4T3,MTA4T4,MTA4T5,MTA5T3,MTA5T4,MTA5T5. 同时为了进一步研究高位MTA近似算法对精度和开销的影响,统一在T5截断处理下设计了MTA6T5和MTA7T5两种高精度近似浮点乘法器.
本节从误差特性和硬件指标2个方面对提出的近似浮点乘法器进行评估,包括提出的设计和现有近似方案,共实现了20种近似浮点乘法器.
为了评估近似乘法器的有效性和量化误差,研究人员制定了多种方法和指标,Liang等人[33]提出的平均相对误差距离(mean of relative error distance,MRED)和归一化平均误差距离(normalized mean error distance,NMED)已被广泛用于评估误差特征. 在n位近似乘法中,MRED统计每组输出的近似结果与精确结果之间的相对偏差;NMED常用于比较不同位宽或规模的近似设计的误差幅度. 以上精度评估指标可用式(8)(9)进行计算.
$ {{MRED}} = \frac{1}{{{2^{2n}}}}\sum\limits_{i = 1}^{{2^{2n}}} {\frac{{|{T_i} - {T'_i}|}}{{{T_i}}}} \text{,}$
$ {{NMED}} = \frac{1}{{{2^{2n}}}}\sum\limits_{i = 1}^{{2^{2n}}} {\frac{{|{T_i} - {T'_i}|}}{{{T_{\max }}}}} \text{,}$
其中,Ti表示精确输出,$T'_i $表示近似输出,Tmax表示最大精确输出.
所有的近似浮点乘法器用Verilog硬件描述语言进行编写,使用Verilator仿真平台进行仿真实验,验证正确性和得到近似结果,而后利用C语言程序计算所有近似乘法器的精确度. 由于单精度浮点数所表示的数据范围较大,因此采用随机生成100万组输入组合的方式计算精度评估指标,如表1所示.
从表1可以看出,与其他近似方案相比,所设计的MRED在2.35×10−3~1.92×10−2之间,具有较小的精度损失. 其中,MTA5T4,MTA4T5~MTA7T5在各项精度指标上均具有显著优势.
硬件资源开销分析与评估实验通过Verilog语言描述乘法器硬件结构,得到相应的RTL代码,然后统一利用EDA工具在28 nm典型工艺下进行逻辑综合. 首先,我们分析和评估了所提出设计的主要电路功能模块的功耗(Power)和面积(Area)开销,如表2所示,以MTA4T5为代表,将32位精确浮点乘法器(32-bit exact floating-point multiplier,EX-FP32)作为对照.
从表2可以看出,所提出设计的MTA和前导“1”检测模块的功耗和面积均较小,在总开销中的占比接近10%,主要的硬件开销来源于精确计算过程和尾数的规格化,实现了降低近似计算成本的目标. 与精确浮点乘法模块相比,尾数乘法模块的功耗下降了91.06%,面积下降了86.87%;规格化模块功耗下降了55.89%,面积下降了25.72%.
除了关注各模块的开销外,整体设计的硬件开销情况是评估的重点,为此,我们在同等条件下实现了包括所提出的设计及其他近似浮点乘法器在内的20种设计,对比了功耗、面积、关键路径延时(Delay)、功耗延时积(PDP)和面积延时积(area-delay product,ADP)等5项指标,如表3所示. PDP、ADP能够分别从功耗和面积与延时方面考虑近似乘法器的硬件资源开销情况,增加对比合理性.
从表3可以看出,与EX-FP32相比,MTA系列乘法器功耗下降83.34%~90.55%,面积下降77.27%~85.80%,PDP下降83.48%~90.69%,ADP下降77.46%~86.01%. 与其他近似方案相比,MTA3T3,MTA4T3和MTA3T4具有最低的功耗和面积开销. 在所提出的MTA系列乘法器设计中,MTA3T3具有最小的开销,MTA7T5具有最大的开销.
综合考虑误差特性和硬件资源开销,分析PDP和ADP分别与MRED的关系,从精度与开销的权衡角度,进一步比较各类近似乘法器的优劣. 不同近似乘法器的硬件指标与MRED的关系如图9所示.
从图9中可以看出,在硬件开销几乎相同的情况下,MTA系列乘法器具有更高的精度,如MTA4T5和MTA5T4,硬件开销与LPC[19]相近,但精度损失却低3~4倍. 另外,本文所提出的大多数近似设计的值位于坐标图的左下角,相比其他近似设计更靠近原点,位于帕累托前沿,表明这些近似设计的精度更高且硬件开销更低. 其中,低位MTA算法相比其他设计虽然硬件开销较低,但是精度损失更大,远离帕累托前沿. 随着MTA算法位宽的增加,当位宽大于6时,精度的提升趋于平缓,综合来看MTA4T5,MTA5T4和MTA5T5在精度和成本之间实现了更好的折衷,使它们成为近似浮点乘法器设计的良好选择.
为了测试和验证本文所提出近似设计的效果,我们进行了有限冲激响应(finite impulse response,FIR)滤波和图像处理的应用实验,通过调用不同的近似浮点乘法器进行相应的仿真测试,并得到相应的测试结果.
FIR滤波器是一种非常稳定的系统,其滤波精度高,可随时修改,不需要考虑阻抗等问题,可以实现特殊要求的幅频特性,在数字信号处理和信号系统等场景中有着广泛的应用.
实验设置中,本文使用等纹波设计法设计了一个15阶的低通滤波器. 设置采样频率Fs为100 MHz,通带频率Fpass为1 MHz,阻带频率Fstop为15 MHz,滤波器的输入信号由2 MHz的正弦波信号和作为噪声的30 MHz的正弦波混合而成. 不同的近似乘法器分别被应用于FIR滤波操作,信噪比(signal-to-noise ratio,SNR)和均方误差(mean square error,MSE)作为评估滤波效果的指标.
我们随机生成了1024组测试数据,并将这些数据作为输入激励,送入由Verilator搭建的仿真环境中. 通过更换不同的近似乘法器得到一系列的测试结果. 最后,对这些结果进行了统计分析,表4列出了不同近似乘法器的SNR和MSE. 为了便于结合硬件资源开销进行分析,将不同近似乘法器的PDP和ADP分别与SNR的关系表示在图10中,从中可以看出所提出的大部分设计位于坐标图的左上角,靠近帕累托前沿,尤其是MTA4T5,MTA5T4和MTA5T5等相比其他设计更靠近左上角,即这些近似乘法器具有更优的精度和硬件开销之间的权衡.
图像处理有较广的动态存储和计算范围,需要用到大量的浮点运算,并且由于人眼难以识别图像的精细变化等原因,使得近似计算可以得到非常好的应用. 在图像处理测试中,峰值信噪比(peak signal-to-noise ratio,PSNR)和结构相似性(structural similarity,SSIM)是广泛用于评价图像质量和相似程度的2种客观指标,经过处理后图像的PSNR反映了处理程序是否令人满意. PSNR接近50 dB代表处理后的图像仅有些许非常小的误差,大于30 dB代表图像质量高,低于30 dB人眼就可以察觉出图像的异常.
为了更清楚地了解各种近似方案在应用中的精度差异,我们分别选择图像自乘和图像融合进行应用测试. 图像自乘能够显著提高图像的对比度,放大近似计算引入的误差. 同时,为了防止单个图像的处理结果出现随机变化,在图像自乘中采用了5张不同的RGB彩色图像进行实验,如图11所示. 实验得到的PSNR和SSIM如表5所示. 为了便于分析精度和硬件开销之间的关系,图12展示了不同近似乘法器的硬件指标与PSNR的关系.
图像融合实验是将2幅不同的灰度图融合并输出为一张图像,本文分别选取了“Walkbridge”דHouse”,“Cameraman”דLake”和“Airplane”דClock”共3组图像进行融合测试,图像融合效果如图13所示,实验得到的PSNR和SSIM如表6所示. 同样,图14展示了不同近似乘法器在图像融合中的PSNR与硬件指标的关系.
实验结果表明,MTA5T4,MTA4T5~MTA7T5的精度优于其他近似乘法器,在图像处理中具有最高的PSNR值和SSIM值. 在图12和图14中,可以看出所提出的大部分乘法器的值位于坐标关系图的左上角,相比其他乘法器更靠近帕累托前沿,这表明所提出的MTA乘法器在精度和硬件开销方面具有更好的权衡,进一步验证了所提出的近似浮点乘法器设计方案在图像处理应用中的有效性和实用性.
本文提出了一种可重构的32位单精度近似浮点乘法器设计方案. 为了进一步优化当前近似乘法器设计的精度和开销之间的平衡,我们提出了MTA的近似计算方法,通过将乘法运算近似为加法和移位运算,可以实现乘法低权重部分的近似计算. 在高有效位进行了多种不同的非对称截断处理并执行精确计算来保证精度和降低开销. 非对称截断配合MTA近似算法实现了近似设计的可重构性,产生了较大的近似设计空间. 与精确浮点乘法器相比,所提出的乘法器功耗下降83.34%~90.55%,面积下降77.27%~85.80%,PDP下降83.48%~90.69%,ADP下降77.46%~86.01%. 与其他先进的近似设计相比,提出的多种乘法器MRED在0.4%以内. 最后,在FIR滤波和图像处理应用中验证了设计的优越性.
作者贡献声明:李鹏程提出了设计优化思路、完成实验并撰写论文;黄立波、陈刚和王永文对设计和实验提出指导意见并进行了论文的修改;赖明澈、邓林、刘威和杨乾明修改了论文.

参考文献 引证文献
排序方式:
1
Al-Fuqaha A, Guizani M, Mohammadi M, et al. Internet of things: A survey on enabling technologies, protocols, and applications[J]. IEEE Communications Surveys & Tutorials, 2015, 17(4): 2347−2376
2
Zhuo Cheng, Luo Shaoheng, Gan Houle, et al. Noise-aware DVFS for efficient transitions on battery-powered IoT devices[J]. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2020, 39(7): 1498−1510
3
丁文隆,汪承宁,童薇. 基于自选尾数压缩的高能效浮点忆阻存内处理系统[J]. 计算机研究与发展,2022,59(3):533−552
Ding Wenlong, Wang Chengning, Tong Wei. Energy efficient floating-point memristive in-memory processing system based on self-selective mantissa compaction[J]. Journal of Computer Research and Development, 2022, 59(3): 533−552 (in Chinese)
4
Liu Weiqiang, Lombardi F, Shulte M. A retrospective and prospective view of approximate computing[J]. Proceedings of the IEEE, 2020, 108(3): 394−399
5
Lundstrom M. Moore’s law forever?[J]. Science, 2003, 299(5604): 210−211
6
Shafique M, Hafiz R, Rehman S, et al. Invited: Cross-layer approximate computing: From logic to architectures[C]//Proc of the 53rd Annual Design Automation Conf. New York: ACM, 2016: 1−6
7
Han Jie, Orshansky M. Approximate computing: An emerging paradigm for energy-efficient design[C]//Proc of IEEE European Test Symp (ETS). Piscataway, NJ: IEEE, 2013: 1−6
8
Deng Jianing, Shi Zhiguo, Zhuo Cheng. Energy-efficient real-time UAV object detection on embedded platforms[J]. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2020, 39(10): 3123−3127
9
Gupta V, Mohapatra D, Park S P, et al. IMPACT: Imprecise adders for low-power approximate computing[C]//Proc of Int Symp on Low Power Electronics and Design. Piscataway, NJ: IEEE, 2011: 409−414
10
Venkataramani S, Chippa V K, Chakradhar S T, et al. Quality programmable vector processors for approximate computing[C]//Proc of the 46th Annual IEEE/ACM Int Symp on Microarchitecture. New York: ACM, 2013: 1−12
11
Liang Jian, Tessier R, Mencer O. Floating point unit generation and evaluation for FPGAs[C]//Proc of Annual IEEE Symp on Field-Programmable Custom Computing Machines. Piscataway, NJ: IEEE, 2003: 185−194
12
Imani M, Masich M, Peroni D, et al. CANNA: Neural network acceleration using configurable approximation on GPGPU[C]//Proc of the 23rd Asia and South Pacific Design Automation Conf (ASP-DAC). Piscataway, NJ: IEEE, 2018: 682−689
13
Imani M, Garcia R, Gupta S, et al. RMAC: Runtime configurable floating point multiplier for approximate computing[C]//Proc of the Int Symp on Low Power Electronics and Design. New York: ACM, 2018: 1−6
14
Zuras D , Cowlishaw M , Aiken A , et al. IEEE Standard for Floating-Point Arithmetic[J]. IEEE Std 754−2008, 2008: 1−70
15
Lin C H, Lin I C. High accuracy approximate multiplier with error correction[C]//Proc of Int Conf on Computer Design (ICCD). Piscataway, NJ: IEEE, 2013: 33−38
16
Bhardwaj K, Mane P S, Henkel J. Power- and area-efficient approximate wallace tree multiplier for error- resilient systems[C]//Proc of 15th Int Symp on Quality Electronic Design. Piscataway, NJ: IEEE, 2014: 263−269
17
Liu Cong, Han Jie, Lombardi F. A low-power, high performance approximate multiplier with configurable partial error recovery[C]//Proc of Design, Automation & Test in Europe Conf & Exhibition (DATE). Piscataway, NJ: IEEE, 2014: 1−4
18
Narayanamoorthy S, Moghaddam H A, Liu Z, et al. Energy-efficient approximate multiplication for digital signal processing and classification applications[J]. IEEE Transactions on Very Large Scale Integration Systems, 2015, 23(6): 1180−1184
19
Li Zhenhao, Lu Zhaojun, Jia Wei, et al. Efficient approximate floating-point multiplier with runtime reconfigurable frequency and precision[J]. IEEE Transactions on Circuits and Systems II: Express Briefs, 2024, 71(7): 3533−3537
20
Yao Shangshang, Zhang Liang, Wang Qiong, et al. An efficient hybrid parallel compression approximate multiplier[C]//Proc of Int Conf on Computer Design (ICCD). Piscataway, NJ: IEEE, 2021: 109−116
21
Kumar U A, Bikki P, Veeramachaneni S, et al. Power efficient approximate multiplier architectures for error resilient applications[C]//Proc of IEEE 19th India Council Int Conf (INDICON). Piscataway, NJ: IEEE, 2022: 1−5
22
Park G, Kung J, Lee Y. Simplified compressor and encoder designs for low-cost approximate radix-4 booth multiplier[J]. IEEE Transactions on Circuits and Systems II: Express Briefs, 2023, 70(3): 1154−1158
23
Sabetzadeh F, Moaiyeri M H, Ahmadinejad M. An ultra-efficient approximate multiplier with error compensation for error-resilient applications[J]. IEEE Transactions on Circuits and Systems II: Express Briefs, 2023, 70(2): 776−780
24
Haider M H, Zhang Hao, Ko S B. Decoder reduction approximation scheme for booth multipliers[J]. IEEE Transactions on Computers, 2024, 73(3): 735−746
25
Tan Hongbing, Tong Gan, Huang Libo, et al. Multiple- mode-supporting floating-point FMA unit for deep learning processors[J]. IEEE Transactions on Very Large Scale Integration Systems, 2023, 31(2): 253−266
26
Tan Hongbing, Zhang Jing, Huang Libo, et al. A multi-level parallel integer/floating-point arithmetic architecture for deep learning instructions[C]//Proc of Euro-Par 2023: Parallel Processing. Berlin: Springer, 2023: 260−274
27
Zhang Hao, Ko S B. Variable-precision approximate floating-point multiplier for efficient deep learning computation[J]. IEEE Transactions on Circuits and Systems II: Express Briefs, 2022, 69(5): 2503−2507
28
Saggese G, Napoli E, Strollo A G M. CFPM: Run-time configurable floating-point multiplier[C]//Proc of the 18th Conf on PhD Research in Microelectronics and Electronics (PRIME). Piscataway, NJ: IEEE, 2023: 173−176
29
Saadat H, Bokhari H, Parameswaran S. Minimally biased multipliers for approximate integer and floating-point multiplication[J]. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2018, 37(11): 2623−2635
30
Imani M, Peroni D, Rosing T. CFPU: Configurable floating point multiplier for energy-efficient computing[C]//Proc of the 54th Annual Design Automation Conf 2017. New York: ACM, 2017: 1−6
31
Jha C K, Walia S, Kanojia G, et al. FPCAM: Floating point configurable approximate multiplier for error resilient applications[C]//Proc of Int Symp on Circuits and Systems (ISCAS). Piscataway, NJ: IEEE, 2021: 1−5
32
Towhidy A, Omidi R, Mohammadi K. On the design of iterative approximate floating-point multipliers[J]. IEEE Transactions on Computers, 2023, 72(6): 1623−1635
33
Liang Jinghang, Han Jie, Lombardi F. New metrics for the reliability of approximate and probabilistic adders[J]. IEEE Transactions on Computers, 2013, 62(9): 1760−1771
2025年第62卷第6期
151
66
0
引用本文
BibTeX
文章信息
doi: 10.7544/issn1000-1239.202550116
  • 接收时间:2025-02-26
  • 首发时间:2026-11-10
  • 出版时间:2025-06-02
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2025-02-26
  • 修回日期:2025-04-15
基金
作者信息
    1国防科技大学计算机学院 长沙 410073
    2先进微处理器芯片与系统重点实验室(国防科技大学) 长沙 410073
    3中山大学计算机学院 广州 510275

通讯作者:

黄立波()
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/10.7544/issn1000-1239.202550116
分享至
全文二维码

扫描看全文

引用本文
BibTeX
本文的引用情况
2种不同金属材料的力学参数
科
Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)
属
Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏