基于四阶统计量的盲源分离算法（它是特征矩阵联合对角化的简称）在特定情况下具有普遍的适用性-Fourth-order statistics-based blind source separation algorithm (which is the characteristic matrix joint diagonalization short) in a particular case of universal applicability

JAD矩阵联合对角化，此数据包包括各种矩阵联合对角化的方法，JADE特征矩阵联合对角化，盲源分离方法的经典方法。有Cardoso开发-JAD，JADE

对接收到得信号进行分离是信号处理过程中经常用到的，然而在完全不知道接受信号先验知识的情况下进行分离就是所谓的盲源分离，此程序就是重多盲源分离算法中比较重要的一种变步长自适应盲源分离算法——EASI-Butt had been received too signal separation is the signal processing is often used, but do not know to accept in full prior knowledge of the signal to

基于峭度的盲源分离算法，包含画图子程序。-Based on Kurtosis algorithm for blind source separation, including the drawing subroutine.

TFBSS是一种基于短时傅里叶时频分析的盲源分离算法，算法基于卷积混合。处理非平稳源信号。-TFBSS performs Blind Source Separation of (over)determined multiplicative mixtures of non-stationary real valued sources. TFBSS is based on the joint-diagonalization of whitened and noise-compensated