黑龙江科技大学学报2023，Vol.33Issue(6)：917-921,5.DOI:10.3969/j.issn.2095-7262.2023.06.022
四次样条插值的改进Fast-ICA算法
Improved Fast-ICA algorithm based on quadruple spline interpolation
郭松林 1丘斯帆1
作者信息
- 1. 黑龙江科技大学 电气与控制工程学院,哈尔滨 150022
- 折叠
摘要
Abstract
This paper aims to improve the performance and accuracy of the blind source separation algorithm.The study involves improving the second-order Newton iteration method of the traditional Fast-ICA algorithm to the fourth-order convergence Newton iteration method with the four-spline interpolation,and constructing four-spline interpolation function;verifying the convergence characteristics of the func-tion through error analysis;simulating and analyze the strength of the original algorithm and the improved algorithm by using Matlab software.The results show that the improved fourth-order convergence Newton iteration method has the fourth-order convergence with the increase of the accuracy and convergence rate by 24%and 35%respectively.关键词
盲源分离/Fast-ICA/四次样条插值Key words
blind source separation/Fast-ICA/quad spline interpolation分类
电子信息工程引用本文复制引用
郭松林,丘斯帆..四次样条插值的改进Fast-ICA算法[J].黑龙江科技大学学报,2023,33(6):917-921,5.
