光纤传感信号特征提取方法研究
首发时间:2009-03-14
摘要:随着光纤技术的不断发展,光纤传感器在安防领域得到了很大的应用,对光纤传感信号的特征提取也成为一个研究热点。本文分析研究了光纤传感信号的三种特征提取方法:MFCC参数特征提取法、小波包能量特征提取法和小波包Shannon熵特征提取法,并且用RBF神经网络分别进行了实验。实验识别结果表明,相比其他两种特征提取方法,小波包Shannon熵特征提取法能产生较好的识别结果。
关键词: 光纤传感信号 特征提取 Mel频率倒谱系数 小波包 Shannon熵
For information in English, please click here
The Research on the Feature Extraction of the Fiber Sensing Signal
Abstract:With the development of the fiber technique, the fiber sensors have been widely applied in the security field and the research on the feature extraction of the fiber sensor signal has become a hot topic. This paper analyzes three methods of the feature extraction of the fiber sensor signal, which take the MFCC, the wavelet packet energy and the wavelet packet Shannon entropy as feature parameter respectively, and does the recognition experiments using RBF neural network. The results show that compared the other two methods, the wavelet packet Shannon entropy feature extraction method can give a better recognition result.
Keywords: fiber sensor signalfeature extraction MFCC wavelet packet Shannon entropy
基金:
论文图表:
引用
No.3027944559812370****
同行评议
共计0人参与
勘误表
光纤传感信号特征提取方法研究
评论
全部评论0/1000