基于自适应评价函数的独立成分分析算法

Adaptive Algorithm for Independent Component Analysis with Flexible Score Functions

简要介绍独立成分分析(ICA)及其模型,然后在极大似然估计的框架下,基于两类参数模型—Gaussian混合密度模型和Pearson系统模型,研究了具有对称分布(包括超高斯分布与亚高斯分布)和非对称分布源混合信号的盲分离问题,给出了一种有效的基于灵活评价函数的ICA新算法,该算法在一定意义上实现了对源信号概率分布的真正全“盲”。与原有的ICA算法相比,该算法具有更广泛应用范围。模拟实验验证了算法的有效性。

After giving a brief introduction about the idea of the model of Independent Component Analysis （ICA）, an algorithm for ICA without any knowledge of their probability distributions was provided. It was achieved under a maximum likelihood framework by considering Gaussian parametric density mixture model and Pearson system model. As a result, an explicit ICA algorithm with flexible score functions to various marginal densities was obtained. Simulation result shows that the proposed algorithm is able to separate a wild range of source signals, including sub-Gaussian and super-Gaussian sources, symmetric and asymmetric sources.