摘要
独立成份分析是信号处理的一项新技术；用来从观测的多维混合信号中提取具有统计独立性的成份。本文基于互信息极小提出了ＩＣＡ的梯度算法。文中通过信号分离测试和与固定点算法的对比以及癫痫脑电特征波提取实验证实了此算法的有效牲。
Independent Component Analysis (ICA) is a new technique in signal processing, which extracts components that are statistically independent to each other from the observed multidimensional mixture signals. In this paper, a simple gradient algorithm of ICA is developed using minimum mutual information. The effectiveness of the algorithm is confirmed by signal separation test and the comparison with fix-point algorithm and epileptic characteristic waves detections.
出处
《信号处理》
CSCD
2001年第6期506-509,532,共5页
Journal of Signal Processing
关键词
独立成份分析
梯度算法
信号分离
信号处理
Independent Component Analysis (ICA) The gradient algorithm Signal separation
