基于HHT及主成分分析的光缆识别信号特征提取  

Feature extraction method of optical cable identification based on HHT and PCA

作　　者：韩晓伟 田晓蓓 HAN Xiaowei;TIAN Xiaobei(The 54th Research Institute of China Electronics Technology Group Corporation(CETC54),Shijiazhuang,Hebei 050081,China)

机构地区：[1]中国电子科技集团公司第五十四研究所,河北石家庄050081

出　　处：《河北工业科技》2022年第5期381-387,共7页Hebei Journal of Industrial Science and Technology

摘　　要：为了解决当环境噪声较严重时光强难以维持稳定,容易误判目标光缆问题,根据干涉信号非线性、非平稳的时域特征,提出了一种希尔伯特-黄变换及主成分分析相结合的信号处理方法。探测器接收的干涉信号对进行经验模态分解,把分解得到的本征模态分量作为列向量构建主成分分析的样本矩阵,以奇异值的贡献率进行有效成分筛选;然后采用主成分分析算法完成降噪优化。最后将降噪后的信号进行希尔伯特-黄变换得到瞬时特征信号,完成目标光缆识别。结果表明,相同噪声环境下所提方法可有效提高回波干涉信号的信噪比。因此,研究结果可以降低误判目标光缆的概率,帮助工程人员更有效地完成光缆识别。When the environmental noise is serious,the light intensity is difficult to remain unchanged in a seriously disturbed environment,which is prone to lead to false positives.As to this problem,considering the non-linear and non-stationary time-domain characteristics of the interference signal,a signal processing method combining Hilbert-Huang transform(HHT)and PCA(Principal Component Analysis)was proposed.The echo interference signal received by the detector was decomposed by EMD to obtain the intrinsic mode components,which were designed as column vectors to construct the sample matrix for PCA.The contribution rate of singular value is considered to screen the principal component,and the PCA algorithm was used to reduce the noise.Finally the de-noised signal was decomposed by HHT to obtain the characteristic signal,and to accomplish the optical cable identification.The results show that the method can effectively improve the signal-to-noise ratio of the echo interference signal at the same noise level.Therefore,the proposed feature extraction method will reduce the false positive of cable identification and help engineers to identify the optical cable more effectively.

关 键 词：数据处理 经验模态分解 希尔伯特-黄变换 主成分分析 光缆识别 

分 类 号：TN98[电子电信—信息与通信工程]