Pulsar Signal Denoising Method Based on Empirical Mode Decomposition and Independent Component Analysis

This paper introduces a novel EMD-ICA denoising method, where relevant modes are selected using a hausdorff distance (HD) measure between the probability density function (pdf) of the input and that of each mode. Then, the first layer intrinsic modal function (IMF) is used to construct multidimensional input vectors. These multidimensional input vectors are used as the input of ICA to realize the filtering of each layer of IMF with aliasing. Numerical simulation was carried out to evaluate the effectiveness of the proposed method. Simulation and test results show its superior performance over other traditional denoising methods in whole.