计算机工程与应用 ›› 2012, Vol. 48 ›› Issue (5): 153-155.

• 数据库、信号与信息处理 • 上一篇    下一篇

五阶收敛的FastICA改进算法

赵天娇,何选森   

  1. 湖南大学 计算机与通信学院,长沙 410082
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2012-02-11 发布日期:2012-02-11

Improved FastICA algorithm with fifth-order convergence

ZHAO Tianjiao, HE Xuansen   

  1. School of Computer and Communication,Hunan University,Changsha 410082,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2012-02-11 Published:2012-02-11

摘要: 目前应用较多的FastICA算法利用了二阶收敛的牛顿迭代法进行优化,为了加快算法的收敛速度,用五阶收敛的牛顿迭代法对其进行改进,得出了两种改进的FastICA算法,通过实验验证了改进算法的性能。

关键词: 独立分量分析, FastICA, 牛顿迭代法, 五阶收敛

Abstract: FastICA algorithm is optimized by the quadratic convergence of Newton iteration method and is widely applied in BBS. In order to accelerate the convergence speed of the algorithm, this paper gives two improved FastICA algorithms which use the fifth-order convergence of Newton iterative method. The experimental results show the performance of the improved algorithms is better than that of the FastICA.

Key words: independent component analysis, FastICA, Newton’s iteration method, fifth-order convergence