数字串行DSP系统的性能权衡

H. Suzuki, Yun-Nan Chang, K. Parhi
{"title":"数字串行DSP系统的性能权衡","authors":"H. Suzuki, Yun-Nan Chang, K. Parhi","doi":"10.1109/ACSSC.1998.751522","DOIUrl":null,"url":null,"abstract":"This paper addresses performance tradeoffs in digit-serial arithmetic architectures for design of dedicated and programmable DSP systems. The advantages and the disadvantages of the digit-serial approach over the bit-parallel approach are discussed in terms of area, latency and power consumption. The addition and the multiplication are chosen for comparison. Digit-serial adders have a significant area advantage over bit-parallel adders. Digit-serial multipliers, however, have less area advantage due to the large number of feedback registers. In addition, critical paths of digit-serial multipliers cannot be reduced significantly compared to low-latency bit-parallel multipliers such as the Wallace tree multiplier. Quantitative performance tradeoffs in design of both programmable and dedicated DSP systems are considered. It is shown that the digit-serial arithmetic is not suitable for programmable DSPs. However, dedicated DSPs can be successfully implemented with the digit-serial arithmetic in limited silicon area.","PeriodicalId":393743,"journal":{"name":"Conference Record of Thirty-Second Asilomar Conference on Signals, Systems and Computers (Cat. No.98CH36284)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1998-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"19","resultStr":"{\"title\":\"Performance tradeoffs in digit-serial DSP systems\",\"authors\":\"H. Suzuki, Yun-Nan Chang, K. Parhi\",\"doi\":\"10.1109/ACSSC.1998.751522\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper addresses performance tradeoffs in digit-serial arithmetic architectures for design of dedicated and programmable DSP systems. The advantages and the disadvantages of the digit-serial approach over the bit-parallel approach are discussed in terms of area, latency and power consumption. The addition and the multiplication are chosen for comparison. Digit-serial adders have a significant area advantage over bit-parallel adders. Digit-serial multipliers, however, have less area advantage due to the large number of feedback registers. In addition, critical paths of digit-serial multipliers cannot be reduced significantly compared to low-latency bit-parallel multipliers such as the Wallace tree multiplier. Quantitative performance tradeoffs in design of both programmable and dedicated DSP systems are considered. It is shown that the digit-serial arithmetic is not suitable for programmable DSPs. However, dedicated DSPs can be successfully implemented with the digit-serial arithmetic in limited silicon area.\",\"PeriodicalId\":393743,\"journal\":{\"name\":\"Conference Record of Thirty-Second Asilomar Conference on Signals, Systems and Computers (Cat. No.98CH36284)\",\"volume\":\"28 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"1998-12-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"19\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Conference Record of Thirty-Second Asilomar Conference on Signals, Systems and Computers (Cat. No.98CH36284)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ACSSC.1998.751522\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Conference Record of Thirty-Second Asilomar Conference on Signals, Systems and Computers (Cat. No.98CH36284)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ACSSC.1998.751522","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 19

摘要

本文讨论了专用和可编程DSP系统设计中数字串行算法体系结构的性能权衡。从面积、延迟和功耗等方面讨论了数字串行方法相对于位并行方法的优缺点。选择加法和乘法进行比较。数字串行加法器比位并行加法器具有显著的面积优势。然而,由于大量的反馈寄存器,数字串行乘法器的面积优势较小。此外,与Wallace树乘法器等低延迟位并行乘法器相比,数字串行乘法器的关键路径不能显著减少。考虑了可编程和专用DSP系统设计中的定量性能权衡。结果表明,数字串行算法不适用于可编程dsp。然而,专用dsp可以在有限的硅面积上成功地实现数字串行算法。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Performance tradeoffs in digit-serial DSP systems
This paper addresses performance tradeoffs in digit-serial arithmetic architectures for design of dedicated and programmable DSP systems. The advantages and the disadvantages of the digit-serial approach over the bit-parallel approach are discussed in terms of area, latency and power consumption. The addition and the multiplication are chosen for comparison. Digit-serial adders have a significant area advantage over bit-parallel adders. Digit-serial multipliers, however, have less area advantage due to the large number of feedback registers. In addition, critical paths of digit-serial multipliers cannot be reduced significantly compared to low-latency bit-parallel multipliers such as the Wallace tree multiplier. Quantitative performance tradeoffs in design of both programmable and dedicated DSP systems are considered. It is shown that the digit-serial arithmetic is not suitable for programmable DSPs. However, dedicated DSPs can be successfully implemented with the digit-serial arithmetic in limited silicon area.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信