RNS processor using moduli sets of the form 2n±1

Author:

Prediger Vanessa1,Bairros Fábio1,Seman Laio Oriel2ORCID,Bezerra Eduardo Augusto1,Pettenghi Héctor1

Affiliation:

1. Universidade Federal de Santa Catarina Florianopolis Brazil

2. Universidade do Vale do Itajai Itajai Brazil

Abstract

SummaryIn recent years, research on residue number systems (RNS) has targeted larger dynamic ranges to explore their inherent parallelism further. In this paper, we start from the traditional 3‐moduli set , with an equivalent 3 ‐bit dynamic range, and propose balanced horizontal extensions by using modulus of the form and to scale the dynamic range and enhance the parallelism according to the requirements. A method to design the reverse conversion by using it is also presented. A case study of multiplier‐accumulator (MACs) in cascade for filtering applications has been analyzed to validate the RNS processor using the proposed moduli set and reverse conversion architecture. Experimental results show that the proposed approaches achieve a significant speedup compared with the state‐of‐the‐art cases for the same dynamic range purposes.

Publisher

Wiley

Subject

Applied Mathematics,Electrical and Electronic Engineering,Computer Science Applications,Electronic, Optical and Magnetic Materials

Reference33 articles.

1. SkavantzosA AbdallahM.Novel residue arithmetic processors for high speed digital signal processing. In: Proc. of the 32nd Asilomar Conf. on Signals Systems and Computers Vol. 1;1998:187‐193.

2. High performance, reduced complexity programmable RNS-FPL merged FIR filters

3. FernandezP GarciaA RamirezJ ParrillaL LlorisA.A RNS‐based matrix‐vector‐multiply FCT architecture for DCT computation. In: Proc of the 43rd IEEE Midwest Symposium on Circuits and Systems Vol. 1;2000:350‐353.

4. RamirezJ GarciaA Meyer‐BaeseU LlorisA.Fast RNS‐FPL based communications receiver design and implementation. In: Proc. Int. Conf. on Field Programmable Logic Vol. 2438;2002:472‐481.

Cited by 2 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3