基于ICA算法的智能电子鼻在混合气体特征提取中的应用

Based on Independent Component Analysis Algorithm of Intelligent Electronic Nose in the Mixed Gas of Feature Extraction

电子鼻传感器在对环境污染的混合气体浓度监测及对工业废气检测中具有重要的作用,但由于现有算法的辨识能力和抗干扰能力差,影响提取原始信息信号的准确度;独立分量分析(ICA)方法是一种高效盲信号分离方法;它将独立的源信号从混合信号中分离出来;文中经过电子鼻传感器检测出混合气体信号,通过ICA算法对混合气体进行分解,对外界干扰噪声进行消除,从而使气体成分辨别达到很好的效果;最后经过MATLAB仿真验证,对辨识出来的原始气体成分具有高精度,强抗干扰能力。

Electronic nose sensors in the gas mixture concentration of environmental pollution monitoring and detection of industrial waste gas has an important role,but because of the recognition capacity of existing algorithms and anti-jamming ability is poor,affecting recognition accuracy.Independent Component Analysis is a highly efficient method of blind signal separation.It an independent source signal from the mixed-signal separation.This paper through the electronic nose sensors to detect gas mixture signal,through the ICA decomposition algorithm of mixed gases on the outside interference to eliminate the noise,so that gas composition identified to achieve good results.Thanks to MATLAB simulation on the identification of the original gas composition come out with high precision.