基于变分模态分解和奇异谱分析的激光雷达信号去噪方法OA北大核心CSTPCD

The LiDAR signals often contain more noise,which not only degrades the signal quality,but also affects the subsequent cloud-aerosol layer detection and aerosol optical thickness inversion.In view of this,a LiDAR signal noise reduction method based on variational mode decomposition(VMD)is improved.The correlations between the mode components and the LiDAR signals are taken into account.The effective signals in the mode components are extracted by the method of Pearson correlation coefficient(PCC).In terms of the low and medium frequency oscillations after the implementation of VMD,secondary filtering is performed with singular spectrum analysis(SSA)to further improve the signal-to-noise ratio(SNR).The effectiveness of the proposed method is verified.The satellite-borne LiDAR noise-containing signals with different SNRs when it is clear or cloudy are simulated.The wavelet transform(WT),the local empirical mode decomposition(EMD),the traditional VMD method and the proposed method are adopted to perform the noise reduction process,and their performances of noise reduction are contrasted.The experimental results show that the improved VMD method can improve the SNR of LIDAR signals effectively.