Geomagnetic signal de-noising method based on improved empirical mode decomposition and morphological filtering

Abstract

To solve the problem that geomagnetic signals are susceptible to random noise and instantaneous pulse interference in geomagnetic navigation, a geomagnetic signal de-noising method based on improved empirical mode decomposition (IEMD) and morphological filtering (MF) is proposed. The instantaneous pulse interference is eliminated by designing different structural elements according to the characteristics of the pulse signal. The signal after filtering the instantaneous pulse interference is decomposed by EMD, and the intrinsic mode functions (IMFs) obtained from the decomposition are determined as two modes (i.e. noise IMFs and mixed IMFs) by the cross-correlation coefficient criterion. The noise IMFs are removed directly, and a normalized least means square filter (NLMS) is designed to remove noise from mixed IMFs, which can adaptively adjust the filtering parameters according to the noise level of different IMF components. The noise-reduced mixed IMFs and residual are reconstructed to obtain the final geomagnetic signal. Experiment results illustrate that the proposed MF-IEMD method can effectively achieve noise reduction. Comparing with the traditional EMD and MF-EMD de-noising methods, the root mean square errors(RMSE) decreased by 49.27% and 24.79%, respectively.

Improved Iterative Closest Contour Point Matching Navigation Algorithm Based on Geomagnetic Vector

“…Secondly, considering that the total geomagnetic features of the matching areas are very similar, the ICCP algorithm is prone to the problem of low matching accuracy or mismatch. And the terrain has only elevation information, while the geomagnetic field has abundant information features, such as vector features, three axes and seven components [15,16]. The objective function is improved by using the geomagnetic three-dimensional features, which is defined as the matching index.…”

Application of IMMF–IHHT algorithm to suppressing random interference of geomagnetic sensors

“…Based on Li Ji's research method, Diao Yun-yun [15] introduced autocorrelation function to verify that the highorder components above order 6 contain effective geomagnetic signal cutoff frequency, identify the effective geomagnetic signal in the medium and high-frequency components of IMF and calculate the root-mean-square error (RMSE) between the final result and the fast Fourier transform result, proving the superiority of morphological filtering and HHT combined filtering. Zhai [16] adopted the improved empirical mode decomposition for geomagnetic measurement signals, designed normalized least mean square (NLMS) filter and morphological filter to remove the noise in the mixed intrinsic mode function (IMF), adaptively adjusted the filtering parameters according to the noise level of different IMF components, reconstructed the IMF and residual after noise reduction and obtained the filtered signal. Zhou [17] used the combined filtering algorithm of adaptive median filter and HHT algorithm for magnetic sensor signal processing and proposed a similarity criterion based on IMF processing.…”