论文部分内容阅读
将多个随机变量的实现的每个分量分别用相应随机变量的分布函数变换 ,再将值域空间划分为均匀格栅 ,则格栅中被占据的格子数目的期望当这些随机变量相互独立时最大。本文发现了这一有趣的规律 ,对其作出了理论解释 ,并据此提出了一种简单的盲源分离方法。该方法能够分离各种源信号且易于用数字硬件实现。
When each component of the implementation of multiple random variables is transformed by the distribution function of the corresponding random variable and then the domain space is divided into uniform grids, the number of grids occupied in the grid is expected. When these random variables are independent of each other maximum. This article finds out this interesting law and makes a theoretical explanation for it. Based on this, a simple blind source separation method is proposed. This method can separate various source signals and is easy to implement with digital hardware.