基于四阶牛顿迭代法的Fast-ICA改进算法  

Improved Fast-ICA algorithm based on fourth order Newton iterative method

作　　者：郭松林 徐海鹏 Guo Songlin;Xu Haipeng(School of Electrical & Control Engineering, Heilongjiang University of Science & Technology, Harbin 150022,China)

机构地区：[1]黑龙江科技大学电气与控制工程学院,哈尔滨150022

出　　处：《黑龙江科技大学学报》2020年第6期698-702,共5页Journal of Heilongjiang University of Science And Technology

摘　　要：为了提高快速独立分量分析算法的收敛速度和分离性能,改进了算法中的二阶牛顿迭代法,使其满足四阶牛顿收敛特性。通过Matlab软件对改进算法进行了仿真实验。结果表明,相比原有二阶收敛算法,改进后算法的误差更小,算法的收敛速度平均提高了23%,分离性能有较大提升。This paper aims to improve the convergence speed and separation performance of the fast independent component analysis algorithm.The study is focused on improving the second-order Newton iterative method to meet the fourth-order Newton convergence characteristics;and verifying the convergence speed and separation performance of the improved algorithm using Matlab software simulation.The results show that:compared with the original second-order convergence algorithm,the improved algorithm demonstrates a smaller the error and an increased convergence speed of the algorithm by an average of 23%,and a greatly improved separation performance.

关 键 词：盲源分离 独立分量分析 Fast-ICA 牛顿迭代法 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]