说明：  本论文的主要工作在于引入了一种新的特征提取方法----独立分量分析。独立分量分析的根本原理是通过分析多维观测数据间的高阶统计相关性，找出相互独立的隐含信息成分，完成分量间高阶冗余的去除及独立信源的提取
(In this paper, the major work is the introduction of a new method of feature extraction independent component analysis. Independent component analysis of the fundamental principle is that through the analysis of multi-dimensional data between the relevance of higher-order statistics, independent of each other to identify the implied message composition, the completion of inter-component higher-order removal of redundant and independent source extraction)