首页> 外文期刊>IEEE Transactions on Circuits and Systems. II, Express Briefs >New bit-serial VLSI implementation of RNS FIR digital filters
【24h】

New bit-serial VLSI implementation of RNS FIR digital filters

机译:RNS FIR数字滤波器的新位串行VLSI实现

获取原文
获取原文并翻译 | 示例

摘要

A bit-serial hybrid VLSI architecture that consists of both table-look-up and conventional binary modules is proposed to implement FIR digital filters using the residue number system (RNS). The architecture is constructed based on a new theorem for performing the operation Α+xΒ mod m without look-up tables, where m is a modulus in the RNS, Α and Β are two numbers in module m, and x∈{0,1}. As compared to a bit-parallel hybrid realization method described recently, the proposed bit-serial one does not need to broadcast input data to the processing elements used and reduces the table-look-up memory in each module processor from (B+2upper bound [log(2N])-1)·B·22B bits to B·22B bits, where B is the wordlength of each modulus and N is the number of filter coefficients. As a consequence, it can provide better performance in VLSI implementation for applications where large moduli and/or large filter orders are used
机译:提出了一种由表查找和常规二进制模块组成的位串行混合VLSI架构,以使用残数系统(RNS)来实现FIR数字滤波器。该体系结构是基于一个新的定理构造的,用于执行无查找表的运算A +xΒmod m,其中m是RNS中的模数,A和Β是模块m中的两个数,并且x∈{0,1 }。与最近描述的位并行混合实现方法相比,所提出的位串行不需要将输入数据广播到所使用的处理元件,并且每个模块处理器中的表查找存储器都从(B + 2 [log(2N])-1)·B·22B位至B·22B位,其中B是每个模数的字长,N是滤波器系数的数量。因此,对于使用大模数和/或大滤波器阶数的应用,它可以在VLSI实现中提供更好的性能。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号