五阶收敛的FastICA改进算法  

Improved FastICA algorithm with fifth-order convergence

作　　者：赵天娇[1] 何选森[1] 

机构地区：[1]湖南大学计算机与通信学院,长沙410082

出　　处：《计算机工程与应用》2012年第5期153-155,共3页Computer Engineering and Applications

基　　金：湖南省湘潭市科技计划项目(No.ZJ20071008)

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。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.

关 键 词：独立分量分析 FASTICA 牛顿迭代法 五阶收敛 

分 类 号：TN911.4[电子电信—通信与信息系统]