独立分量分析及其在地震信息处理中应用初探

独立分量分析(ICA)是最近才发展起来的一种统计学方法，旨在寻求对非高斯分布数据进行有效表示，使得各个分量在统计学上独立，或者尽最大可能地独立。许多应用中，这种表示意在获取数据的基本结构，包括特征提取和信号分离。本文给出ICA的基本理论和快速算法，并对FastICA稍作改进，在分析地震信号特点的基础上，对其在地震信息处理中应用进行初步探索，表明ICA在地震信号处理中具有应用前景。

Independent component analysis and its testing application on seismic signal processing

ICA is a novel statistical method developed recently, which is used to find a representation of the Non Gaussian multivariate data, so that each components of the vector are independent statistically, or as independent as possible. In many applications, the transformation aims to capture the basic structure in data, including features abstraction and signals separation. In this paper, we present the principal theory and fast algorithms, at the same time, realizing the FastICA and its slightly updating. On the base of analyzing the features of seismic signals, we do preliminary studies and try to apply ICA on seismic signal processing. Our works show the good perspective of ICA application on seismic signal processing.