一种新的基于峰度的盲源分离开关算法（附参考文章），无需假设源信号的概率密度函数, 可直接对独立分量分析中的激活函数进行自适应学习。-A New kurtosis switching algorithm for blind source separation (with reference to the article), without assuming that the probability density function of the source signal, independent component analysis can activate the function of direct adaptive learning.