Assertion Driven Modified Booth Encoding and Post Computation Model for Speed MAC Applications

Sivasaravanababu, T. R. Dineshkumar, Dr. G. Saravana Kumar
{"title":"Assertion Driven Modified Booth Encoding and Post Computation Model for Speed MAC Applications","authors":"Sivasaravanababu, T. R. Dineshkumar, Dr. G. Saravana Kumar","doi":"10.3233/apc210289","DOIUrl":null,"url":null,"abstract":"The Multiply-Accumulate Unit (MAC) is the core computational block in many DSP and wireless application but comes with more complicated architectures. Moreover the MAC block also decides the energy consumption and the performance of the overall design; due to its lies in the maximal path delay critical propagation. Developing high performance and energy optimized MAC core is essential to optimized DSP core. In this work, a high speed and low power signed booth radix enabled MAC Unit is proposed with highly configurable assertion driven modified booth algorithm (AD-MBE). The proposed booth core is based on core optimized booth radix-4 with hierarchical partial product accumulation design and associated path delay optimization and computational complexity reduction. Here all booth generated partial products are added as post summation adder network which consists of carry select adder (CSA) & carry look ahead (CLA) sequentially which narrow down the energy and computational complexity. Here increasing the operating frequency is achieved by accumulating encoding bits of each of the input operand into assertion unit before generating end results instead of going through the entire partial product accumulation. The FPGA implementation of the proposed signed asserted booth radix-4 based MAC shows significant complexity reduction with improved system performance as compared to the conventional booth unit and conventional array multiplier.","PeriodicalId":429440,"journal":{"name":"Recent Trends in Intensive Computing","volume":"74 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Recent Trends in Intensive Computing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.3233/apc210289","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

The Multiply-Accumulate Unit (MAC) is the core computational block in many DSP and wireless application but comes with more complicated architectures. Moreover the MAC block also decides the energy consumption and the performance of the overall design; due to its lies in the maximal path delay critical propagation. Developing high performance and energy optimized MAC core is essential to optimized DSP core. In this work, a high speed and low power signed booth radix enabled MAC Unit is proposed with highly configurable assertion driven modified booth algorithm (AD-MBE). The proposed booth core is based on core optimized booth radix-4 with hierarchical partial product accumulation design and associated path delay optimization and computational complexity reduction. Here all booth generated partial products are added as post summation adder network which consists of carry select adder (CSA) & carry look ahead (CLA) sequentially which narrow down the energy and computational complexity. Here increasing the operating frequency is achieved by accumulating encoding bits of each of the input operand into assertion unit before generating end results instead of going through the entire partial product accumulation. The FPGA implementation of the proposed signed asserted booth radix-4 based MAC shows significant complexity reduction with improved system performance as compared to the conventional booth unit and conventional array multiplier.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
高速MAC应用的断言驱动改进摊位编码和后期计算模型
乘法累加单元(MAC)是许多DSP和无线应用的核心计算块,但具有更复杂的架构。此外,MAC块还决定了整体设计的能耗和性能;由于其存在于最大路径上延迟临界传播。开发高性能、节能的MAC核是优化DSP核的必要条件。在这项工作中,提出了一种高速低功耗的签名展位基数支持MAC单元,该单元采用高度可配置的断言驱动改进展位算法(AD-MBE)。提出的展位核心是基于核心优化展位基数-4,采用分层部分产品累积设计和相关的路径延迟优化和计算复杂度降低。在这里,所有生成的部分产品都被添加到由进位选择加法器(CSA)和进位预判加法器(CLA)组成的后求和加法器网络中,从而缩小了能量和计算复杂度。在这里,增加操作频率是通过在生成最终结果之前将每个输入操作数的编码位累加到断言单元中来实现的,而不是通过整个部分乘积累加。与传统的展台单元和传统的阵列乘法器相比,所提出的基于签名断言展台基数4的MAC的FPGA实现显示出显著的复杂性降低和系统性能的改善。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Enhancing Agricultural Product Supply Chain Management Using Blockchain Technology: Concept An Image Encryption Technique Based on Logistic Sine Map and an Encrypted Image Retrieval Using DCT Frequency Study on SDN with Security Issues Using Mininet Smart Ambulance System with Remote Knowledge Communications Through Cloud Diabetes Risk Forecasting Using Logistic Regression
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1