Journal of Systems Engineering and Electronics ›› 2010, Vol. 32 ›› Issue (11): 2304-2309.doi: 10.3969/j.issn.1001-506X.2010.11.10

• 电子技术 • 上一篇    下一篇

基于CORDIC的滑窗最小二乘递推算法

石斌斌,钱林杰,程翥,皇甫堪   

  1. 国防科学技术大学电子科学与工程学院, 湖南 长沙 410073
  • 出版日期:2010-11-23 发布日期:2010-01-03

Sliding window RLS method based on CORDIC

SHI Bin-bin,QIAN Linjie, CHENG Zhu, HUANGFU Kan   

  1. School of Electronic Science and Engineering, National Univ. of Defense Technology, Changsha 410073, China
  • Online:2010-11-23 Published:2010-01-03

摘要:

针对最小二乘算法涉及对数据自相关矩阵求逆,若直接求逆则计算量比较大,且求逆过程对误差比较敏感,有限字长效应明显的问题,提出一种基于Givens旋转的滑窗RLS实现方法,并且给出了CORDIC实现的方法,提高了基于QR分解的最小二乘算法实现的计算效率和数据吞吐率,使其适应于需要极高采样率的应用,并且该算法能够以滑窗形式的样本作为输入,比传统的Givens算法能够适应更多的应用环境。仿真结果验证了该方法的有效性。

Abstract:

A sliding window RLS method based on givens rotation is proposed, which can be used to improve calculative efficiency of the least square adaptive algorithm based on QR decomposition and increase data throughput. This method is suitable for high sampling rate application, and the input data with sliding window form can suit more application environment than traditional Givens algorithm. Simulation results are given to demonstrate the effectiveness of the new method.