中文摘要：

为消除地磁数据中的噪声和干扰，提高地磁匹配导航的精度和可靠性，采用小波降噪方法对地磁数据进行预处理。通过对小波母函数特性的研究，结合地磁场自身规律提出了地磁数据降噪中小波母函数的选取准则，并根据地磁匹配导航的特点提出了基于误差和相关系数的小波分解层数确定方法。通过对小波降噪主要方法的定性分析，采用了小波域软闽值降噪方法。选取4个小波母函数进行了地磁数据降噪和匹配导航仿真实验，实验结果表明小波域软阈值降噪方法是可行有效的，且sym4小波母函数的综合性能最优。

英文摘要：

To eliminate noise and magnetic disturbance from surveyed data of geomagnetic field and improve the accuracy and reliability of geomagnetic matching navigation, a denoising method based on wavelet transform is proposed for pretreatment.Selecting rules of generating wavelets are given after researching the characteristics of generating wavelets and the mechanism of geomagnetic field,and a method to determine the number of decomposition layers for wavelet analysis based on errors and correlative coefficients is also put forward according to the demands of geomagnetic matching navigation.The denoising method based on soft-threshold is adopted through the qualitative analysis of the main denoising methods in wavelet field,and four generating wavelets are used in simulation experiments of geomagnetic data denosing and geomagnetic matching navigation.Simulation results show that the proposed method is feasible and effective, and sym4 is the optimal one of the four wavelets used in this paper for its comprehensive performances.