组合GNSS监测信号降噪方法研究  

Research on Noise Reduction Method of Combined GNSS Monitoring Signals

作　　者：于长龙 沈晶 孙五斌 YU Changlong;SHEN Jing;SUN Wubin(Zhejiang Provincial Institute of Surveying and Mapping Science and Technology,Hangzhou 311100,China)

机构地区：[1]浙江省测绘科学技术研究院,浙江杭州311100

出　　处：《测绘与空间地理信息》2024年第11期145-148,共4页Geomatics & Spatial Information Technology

摘　　要：针对传统全球导航卫星系统(Global Navigation Satellite System,GNSS)监测数据信噪难以分离的问题,综合奇异值分解(Singular Value Decomposition,SVD)模型、完备集合经验模态分解(Complete Ensemble Empirical Mode Decomposition with Adaptive Noise,CEEMDAN)方法、KL散度法(Kullback-Leibler Divergence,KLD)在数据处理中的优势,提出了一种基于SVD-CEEMDAN和KLD的降噪新方法。首先,使用SVD模型对原始信号进行初步去噪,滤除原始信号中的大部分噪声;其次,使用CEEMDAN方法对初步降噪后所得信号进行自适应分解,得到若干个本征模太函数(Intrinsic Mode Function,IMF)并使用KLD选取包含有效信号成分的IMF分量;最后重构有效信号IMF分量得到最终降噪结果。使用仿真信号与桥梁实测GNSS信号对本文提出方法进行实验,结果表明本文提出降噪方法降噪结果的信噪比比单一SVD模型降噪结果更高,均方根误差更小,具有更优的降噪效果,验证了本文方法在实际工程项目中应用的可行性与优越性。This paper aims at the problem that it is difficult to separate the signal and noise of traditional Global Navigation Satellite System(GNSS) monitoring data,and integrates the advantages of Singular Value Decomposition(SVD) model,Complete Ensemble Empirical Mode Decomposition with Adaptive Noise(CEEMDAN) method and Kullback-Leibler divergence(KLD) method in data processing,a new noise reduction method based on SVD-CEEMDAN and KLD is proposed.Firstly,we use SVD model to preliminarily denoise the original signal and filter out most of the noise in the original signal;Secondly,CEEMDAN method is used to adaptively decompose the signal after preliminary noise reduction to obtain several intrinsic mode functions(IMF),and KLD is used to select the IMF component containing effective signal components;Finally,the IMF component of the effective signals is reconstructed to obtain the final noise reduction results.The simulation signal and measured bridge GNSS signal are used to verify the method proposed in this paper.The results show that the signal-to-noise ratio of the noise reduction results by the noise reduction method proposed in this paper is higher than that of the single SVD model,and the root mean square error is smaller,which has a better noise reduction effect,which verifies the feasibility and superiority of the application of the method in actual engineering projects.

关 键 词：奇异值分解 完备集合经验模态分解 KL散度法 降噪 全球导航卫星系统 

分 类 号：P228.4[天文地球—大地测量学与测量工程]