一种新的高速自适应滤波的脉动实现结构

尚勇; 吴顺君; 项海格

一种新的高速自适应滤波的脉动实现结构

尚勇^{①;吴顺君},
吴顺君,
项海格

计量
- 文章访问数: 2203
- HTML全文浏览量: 98
- PDF下载量: 573
- 被引次数: 0
出版历程
- 收稿日期: 2000-12-27
- 修回日期: 2001-04-23
- 刊出日期: 2002-08-19

A novel systolic implementation of the high-speed adaptive filter

Shang Yong^{①;吴顺君},
Wu Shunjun,
Xiang Haige

摘要

摘要: LMS算法具有计算简单,易于实现的特点,被广泛应用于诸如通信和雷达等许多信号处理领域,对其高速实现结构的研究一直是滤波器结构设计中的一个研究重点和热点。该文基于并行流水线LMS(PIPLMS)算法,设计了一种高速自适应滤波器脉动结构。该结构既具有脉动结构的高度流水特性,又具有一定的并行性。与已有结构相比,该文设计的结构具有更高的数据吞吐率。同时由于其并行特性,该结构还具有更低的系统功耗,更大的步长因子选择范围和更快的收敛速度。
- 并行流水线 LMS算法; 脉动阵; 自适应滤波
Abstract: The LMS algorithm is widely used in various signal processing fields such as radar, communication, etc. due to its simplicity. It is a focus in the research on the implementation of the high-speed adaptive filter. Based on the parallel pipelined LMS (PIPLMS) algorithm, a high-speed filter implementation is proposed in this paper. This implementation has both the properties of high pipelining in the systolic structure and parallelism in the parallel structure. At the same time, the implementation presented herein also has the properties of higher throughput rate, lower power consumption, larger step size range maintained convergence behavior and faster convergence speed.

HTML全文

参考文献(1)

M.D. Meyer, D. P. Agrawall, A high sampling rate delayed LMS filter architecture, IEEE Trans.on Circuits and Systems- Ⅱ, 1993, CAS-II-40(11), 727-729.[2]H. Herzberg, R. H. Cohen, Y. Beery, A systolic array realization of an LMS adaptive filter and the effects of delayed adaptation, IEEE Trans. on Signal Processing, 1992, SP-40(11), 2799 2803.[3]J. Thomas, Pipelined systolic architectures for DLMS adaptive filtering, Journal of VLSI Signal Processing 1996, 12(3), 223-246.[4]S.C. Douglas, Q. Zhu, K. F. Smith, A pipelined LMS adaptive FIR filter architecture without adaptation delay, IEEE Trans. on Signal Processing, 1998, SP-46(3), 775 779.[5]N.R. Shanbhag, M. Goel, Low-power adaptive filter architectures and their application to 51.84 Mb/s ATM-LAN, IEEE Trans. on Signal Processing, 1997, SP-45(5), 1276-1290.[6]Q. Zhu, S. C. Douglas, K. F. Smith, A pipelined architecture for LMS adaptive FIR filter without adaptation delay, ICASSP97, Munich, Germany, 1997, 1933 1936. ［7］K. Matsubara, K. Nishikawa, H. Kiya, A new pipelined architecture of the LMS algorithm without degradation of convergence characteristics, ICASSP97, Munich, Germany, 1997, 4125-4128.[7]K. Matsubara, K. Nishikawa, H. Kiya, Pipelined LMS adaptive filter using a new look-ahead transformation, IEEE Trans. on Circuits and Systems-II, 1999, CAS-Ⅱ-46(1), 51 55.[8]尚勇,刘卫东,吴顺君,基于并行算法的自适应滤波原理及其性能分析,电子学报,2001,29(9),1230-1232[9]尚勇,吴顺君,项海格,并行流水线LMS算法及其性能分析,自然科学进展,2001,11(10),1032-1037.[10]N.R. Shanbhag, K. K. Parhi, Relaxed look-ahead pipelined LMS adaptive filters and their application to ADPCM coder, IEEE Trans. on Circuits and Systems-Ⅱ, 1993, CAS-Ⅱ-40(12), 753-766.[11]D.A. Parker, K. K. Parhi, Low-area/power parallel FIR digital filter implementations, Journal of VLSI Signal Processing, 1997, 17(1), 75-92.

施引文献

资源附件(0)

访问统计