点击次数：

论文类型：会议论文

第一作者：Wei, Guangfen

合写作者：Tang, Zhenan,Chan, Philip C.H.,Yu, Jun

发表时间：2004-08-19

收录刊物：EI

卷号：3173

页面范围：696-701

摘要：Blind Source Separation (BSS) has been a strong method to extract the unknown independent source signals from sensor measurements which are unknown combinations of the source signals. In this paper, a BSS based modeling method is proposed and analyzed for a micro gas sensor array, which is fabricated with surface micromachining technology and is applied to detect the gas mixture of CO and CH4. Two widely used BSS methods--Independent Component Analysis (ICA) and Nonlinear Principal Component Analysis (NLPCA) are applied to obtain the gas concentration signals. The analyzing results demonstrate that BSS is an efficient way to extract the components which corresponding to the gas concentration signals. © Springer-Verlag Berlin Heidelberg 2004.