Scholarlink

<p>Empirical Mode Decomposition (EMD) is an adaptive algorithm for decomposing non-stationary data into multiple intrinsic mode functions (IMFs). However, the presence of spurious components and mode mixing in EMD applications often undermines the physical interpretability of IMFs, with their underlying causes not fully understood. Building on existing research, this study further explores the effects of low sampling frequency, noise, signal amplitude, and frequency on EMD. Experimental results show that the occurrence of spurious components depends on the ratio of signal frequency to sampling frequency, while their amplitude is influenced by signal amplitude. Mode mixing is found to be related to the signal-to-noise ratio, sampling frequency, signal frequency ratio, and amplitude ratio, with specific criteria provided. Based on these findings, an improved EMD is proposed by equalizing the amplitude-frequency characteristics of the signal, and its performance is shown to be superior to that of existing EMD variants in experimental examples.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Research and Improvement Methods for EMD on the Basis of Statistical Models

  • 期刊论文

  • 发表日期：

    2025年4月27日

查看全文
  • Jie Fu,
  • Ling Yang,
  • Xiaoqiong Zhen

摘要

Empirical Mode Decomposition (EMD) is an adaptive algorithm for decomposing non-stationary data into multiple intrinsic mode functions (IMFs). However, the presence of spurious components and mode mixing in EMD applications often undermines the physical interpretability of IMFs, with their underlying causes not fully understood. Building on existing research, this study further explores the effects of low sampling frequency, noise, signal amplitude, and frequency on EMD. Experimental results show that the occurrence of spurious components depends on the ratio of signal frequency to sampling frequency, while their amplitude is influenced by signal amplitude. Mode mixing is found to be related to the signal-to-noise ratio, sampling frequency, signal frequency ratio, and amplitude ratio, with specific criteria provided. Based on these findings, an improved EMD is proposed by equalizing the amplitude-frequency characteristics of the signal, and its performance is shown to be superior to that of existing EMD variants in experimental examples.