【摘要】：為解決EMD-IT去噪算法中閾值難以確定的問題,提出一種基于高斯白噪聲能量分布的閾值估計方法。將含噪信號進行經(jīng)驗?zāi)B(tài)分解并估計各固有模態(tài)函數(shù)(IMF)中噪聲的能量;根據(jù)模態(tài)單元閾值的含義,在各IMF中利用去除掉的模態(tài)單元包含的總能量等于噪聲能量這一準則估計閾值。合成數(shù)據(jù)和實際心電信號的去噪仿真實驗驗證了該方法的有效性,其是自適應(yīng)的且避免了閾值選擇的主觀性。
[Abstract]:In order to solve the problem that threshold is difficult to determine in EMD-IT denoising algorithm, a threshold estimation method based on Gao Si white noise energy distribution is proposed. The noised signal is decomposed into empirical mode and the energy of noise in each inherent mode function (IMF) is estimated. The threshold value is estimated by using the criterion that the total energy contained by the removed modal element is equal to the noise energy in each IMF. The simulation results of synthetic data and real ECG signal demonstrate the effectiveness of the proposed method, which is adaptive and avoids the subjectivity of threshold selection.
【作者單位】： 華北科技學院基礎(chǔ)部;中國地質(zhì)大學數(shù)理學院;
【基金】：國家自然科學基金項目(61071188) 中央高校基本科研業(yè)務(wù)費基金項目(3142013026;3142014127) 華北科技學院應(yīng)用數(shù)學重點學科基金項目(HKXJZD201402)
【分類號】：TN911.4
，

本文編號：2246894