[Study on Electrocardiogram Signal De-noising Methods Based on Ensemble Empirical Mode Decomposition Decomposed by...

Area of Science:

• Signal Processing
• Biomedical Engineering
• Data Analysis

Background:

• Ensemble Empirical Mode Decomposition (EEMD) is crucial for analyzing non-stationary signals like electrocardiogram (ECG).
• EEMD's precision is often limited by experience-based, non-adaptive parameter settings (noise ratio, ensemble number).
• These limitations hinder accurate analysis of uncertain signals.

Purpose of the Study:

• To develop an adaptive method for optimizing EEMD parameters.
• To enhance the precision and correctness of ECG signal denoising using EEMD.
• To overcome the limitations of empirical parameter selection in EEMD.

Main Methods:

• Proposed a novel method utilizing white noise decomposed by EEMD.
• Applied Empirical Mode Decomposition (EMD) to segment signals into intrinsic mode functions (IMFs).
• Selected white noise IMFs based on constant energy density and average period product; adaptively determined EEMD parameters to prevent modal aliasing.

Main Results:

• The proposed method adaptively determined EEMD parameters, enhancing signal analysis.
• Successfully utilized white noise characteristics for parameter optimization.
• Demonstrated effective ECG signal denoising through experimental validation.

Conclusions:

• The novel adaptive approach significantly improves EEMD performance for ECG signal denoising.
• This method offers a more robust and adaptable solution compared to traditional empirical parameter settings.
• The findings highlight a promising direction for advanced non-stationary signal processing.