基于投影方法的约束独立成分分析

Constrained Independent Component Analysis Based on Projection Methods

独立成分分析是解决盲源分离问题的一种有效工具,但ICA具有伸缩(dilation)与排序(permutation)的不确定性的本质特征。本文利用一些约束条件,采用Lagrange乘子法并结合简单的投影方法,可以以特定的形式来进行独立成分的排序,并且可以在信号分离过程中规范化解混矩阵(demixingmatrix),能够系统地减轻ICA对于伸缩与排序的不确定性。仿真结果证实了算法的有效性。

As an important technique, independent component analysis (ICA) has been widely applied to blind source separation. But ICA has an inherent indeterminacy on dilation and permutation. In this paper, some constraints can be introduced into ICA, then projection methods and Lagrange multiplier methods are used to order the independent components in a specific manner and normalize the demixing matrix in the signal separation procedure. This can systematically eliminate the indeterminacy of ICA on permutation and dilation. The validity of the algorithms are confirmed by the experiments and results.