经验模态分解在地磁匹配导航中的研究

Empirical Mode Decomposition Method for Geomagnetic Matching Navigation

实测的地磁数据易受到软硬磁干扰,导致地磁数据特征不明显,严重影响地磁匹配导航的准确性。该文对实测的地磁信号分别经小波强制降噪、小波阈值降噪和基于经验模态分解的阈值滤波,处理后的数据与实测的地磁数据分别进行了地磁匹配,匹配结果显示,基于经验模态分解的阈值滤波方法的匹配误差最低。该方法运用在使用磁通门与陀螺组合方式进行的车辆导航中结果表明该方法可以有效提高导航的精度。

The geomagnetic matching navigation which uses the magnetic measurement transducer, is always affected by various noises and disturbances, which will cause the feature of the magnetic indistinct, even make the navigation failure. Thus, it is essential to find a way to improve the geomagnetic accuracy. To make a good contrast, we make the matching navigation experiment by using four kinds of data： the original measurement data, the wavelet forced de-noising data, wavelet thresholding de-noising data and the data which is based on the empirical mode decomposition （EMD） method processing. The result illustrates that the EMD method achieves a good contribution to the geomagnetic matching navigation, it can decompose the signal and remove the noise by the feature of the signal itself, and raise the accuracy of the geomagnetic matching navigation.