摘要
学习速率的优选问题是自适应ICA算法中一个重要问题。论文建立了学习速率与相依性测度之间的一种非线性函数关系,以此为基础提出了一种新的变学习速率的自适应ICA算法,并且分析了参数a,b的取值原则及对算法收敛性能和稳态性能的影响。该算法能根据相依性测度所反映的信号分离的状态自适应地调节学习速率,克服了传统算法在稳态阶段步长调整过程中的不足。理论分析和计算机仿真结果都验证了算法的收敛性能和稳态性能。
An important problem in adaptive ICA algorithm is opting learning rate.This paper establishes a nonlinear functional relationship between learning rate and measure of signal dependence.On the basis of the functional relationship,the author presents the new algorithm of variable learning rate adaptive ICA and analyses convergence and steady of the algorithm with variable a and b.This algorithm can adjust the learning rate on the basis of the degree of signal separation which is reflected by measure of signal dependence,and overcomes the disadvantages of traditional algorithms in the process of step size change of adaptive steady state.The theoretical analysis and computer simulation results show this algorithm has better performance in convergence and steady.
出处
《计算机工程与应用》
CSCD
北大核心
2007年第1期53-56,82,共5页
Computer Engineering and Applications
基金
国家自然科学基金资助项目(60472062)
湖北省自然科学基金资助项目(2004ABA038)。
关键词
盲信号分离
独立分量分析
学习速率
相依性测度
Blind Source Separation（BSS）
Independent Component Analysis（ICA）
learning rate
