Research on combined noise reduction of GNSS elevation data of paddy field grader with EMD and S-G filter

Objective In order to reduce the multipath effect and the influence of machine vibration when measuring terrain elevation information based on the global navigation satellite system (GNSS) paddy field grader, improve the accuracy of the establishment of the datum in front of the paddy field leveling, and then improve the leveling quality.

Method Combining the respective advantages of empirical mode decomposition (EMD) and Savitzky-Golay (S-G) filter, a combined noise reduction method of EMD and S-G was proposed. The method first used EMD to decompose the raw elevation signal into several intrinsic mode functions (IMFs), and used the normalized autocorrelation function and correlation coefficient to subdivide them into noise IMFs, mixed IMFs and effective IMFs, and the S-G algorithm was used to process the mixed IMFs, and finally the S-G filtered IMFs and the effective IMFs were reconstructed to obtain the final denoised data.

Result The static verification test results showed that the root mean square error after combined filtering was 36.9% lower than that before filtering, and the signal to noise ratio was 6.3% higher than that before filtering. The field measurement test results showed that the data fluctuation range after filtering was reduced by 11.9%.

Conclusion The EMD and S-G combined noise reduction algorithm proposed in this paper effectively reduces the multipath effect error and vibration error, improves the smoothness of data, and has practical significance for improving the quality of paddy field leveling operations.