Skip to contentScholarGate
LibraryBookshelfDeskReview StudioAssistant
Sign in
On this page
IntuitionHow it worksWhen to use itStrengths & limitationsCommon pitfallsApplicationsFrequently asked🔒 Read the full methodSourcesRelated methods
Cite this pageSpotted an issue on this page? Report or suggest a fix →
Home›Time Series›CEEMDAN
Process / pipelineEnsemble decomposition

CEEMDAN

Complete Ensemble Empirical Mode Decomposition with Adaptive Noise · Also known as: CEEMDAN, Ensemble EMD with noise

Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) is an improved variant of empirical mode decomposition (EMD) that addresses mode-mixing artifacts through ensemble averaging with adaptive noise. Introduced by Torres and colleagues (2011), CEEMDAN decomposes signals into intrinsic mode functions (IMFs) representing oscillations at different scales. The method adds controlled noise to multiple realizations and averages the results, producing more stable, physically meaningful components than standard EMD.

ScholarGate
  1. Process / pipeline
  2. v1
  3. 3 Sources
  4. PUBLISHED
Cite this page →
Tools & resources
Download slides
Learn & explore

Read the full method

Members only

Sign in with a free account to read this section.

Sign in

Method map

The neighbourhood of related methods — select a node to explore.

CEEMDAN
Empirical Mode Decomposi…Empirical Wavelet Transf…Variational Mode Decompo…

When to use it

Apply CEEMDAN when you need a robust decomposition of non-stationary signals and want to avoid mode-mixing artifacts. It is particularly useful for fault diagnosis in machinery, biomedical signal analysis, financial time series, and any domain where intrinsic oscillations must be separated clearly. CEEMDAN is superior to standard EMD when dealing with complex, multi-component signals.

Strengths & limitations

Strengths
  • Significantly reduces mode-mixing compared to standard EMD, producing more stable and interpretable IMFs
  • Ensemble averaging enhances robustness to noise and transient disturbances
  • Preserves physical meaning of components, making them suitable for feature extraction
  • Works without prior specification of frequency bands or basis functions
Limitations
  • Computationally expensive—requires running EMD hundreds of times, making it slower than wavelet-based methods
  • Requires tuning of noise level and ensemble size; inappropriate choices degrade decomposition quality
  • Like all EMD variants, definition of extrema (local maxima and minima) can be ambiguous at signal boundaries
  • The presence of residual noise in IMFs can complicate interpretation

Frequently asked

What is the difference between CEEMDAN and EEMD?

EEMD adds noise to the original signal and averages the IMFs. CEEMDAN adds noise at each decomposition stage and averages the mode-wise residuals. This adaptive approach in CEEMDAN produces better separation of modes and less mode-mixing than EEMD.

How do I choose the noise level for CEEMDAN?

Noise level is typically expressed as a fraction (0.1 to 0.5) of the signal's standard deviation. Start with 0.2 and adjust empirically: higher noise levels reduce mode-mixing but may blur components; lower levels risk mode-mixing but sharpen boundaries.

How many ensemble members do I need?

At least 100 for reasonable stability; 200–500 is common in research. More members increase computational cost but improve consistency. Use cross-validation or stability analysis to justify your choice.

Can CEEMDAN handle non-stationary signals with trends?

Yes, CEEMDAN extracts the trend as the final residual. However, if a strong polynomial trend dominates, detrending the signal first (e.g., using differentiation or detrending splines) can improve the quality of IMF extraction.

Are CEEMDAN IMFs always invertible?

Yes, by design. The sum of all IMFs plus the final residual reconstructs the original signal exactly. This invertibility is valuable for reconstructing filtered signals or for synthesis applications.

Sources

  1. Torres, M. E., Colominas, M. A., Schlotthauer, G., & Flandrin, P. (2011). A complete ensemble empirical mode decomposition with adaptive noise. In 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 4144–4147). DOI: 10.1109/ICASSP.2011.5947265 ↗
  2. Colominas, M. A., Schlotthauer, G., & Torres, M. E. (2014). Improved complete ensemble empirical mode decomposition with adaptive noise. IEEE Transactions on Signal Processing, 63(6), 1408–1413. link ↗
  3. Huang, N. E., et al. (1998). The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis. Proceedings of the Royal Society of London A, 454(1971), 903–995. DOI: 10.1098/rspa.1998.0193 ↗

How to cite this page

ScholarGate. (2026, June 3). Complete Ensemble Empirical Mode Decomposition with Adaptive Noise. ScholarGate. https://scholargate.app/en/time-series/ceemdan

Related methods

Empirical Mode DecompositionEmpirical Wavelet TransformVariational Mode Decomposition

Which method?

Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.

  • Empirical Mode DecompositionSignal Processing↔ compare
  • Empirical Wavelet TransformTime Series↔ compare
  • Variational Mode DecompositionSignal Processing↔ compare
Compare side by side →

Similar methods

Empirical Mode DecompositionHilbert-Huang TransformVariational Mode DecompositionEmpirical Wavelet TransformSignal DenoisingSynchrosqueezing TransformDiscrete Wavelet TransformMODWT

Related reference concepts

EM AlgorithmEnsemble Forecasting and PredictabilityStochastic OptimizationEnsemble MethodsData AssimilationFourier Transform (Applied)

Spotted an issue on this page? Report or suggest a fix →

ScholarGate — CEEMDAN (Complete Ensemble Empirical Mode Decomposition with Adaptive Noise). Retrieved 2026-07-21 from https://scholargate.app/en/time-series/ceemdan · Dataset: https://doi.org/10.5281/zenodo.20539026
Quick facts
Originator
María E. Torres
Subfamily
Ensemble decomposition
Year
2011
Type
Non-stationary signal decomposition
Related methods
Empirical Mode DecompositionEmpirical Wavelet TransformVariational Mode Decomposition
ScholarGate

A content-first reference library for research methods — what each one is, how it works, and where it comes from.

Open data (CC-BY)

Explore

  • Library
  • Search the library…
  • Browse by field
  • Fields
  • Journey
  • Compare
  • Which method?

Reference

  • Subjects
  • Atlas
  • Glossary
  • Methodology
  • Philosophy

Your tools

  • Bookshelf
  • Desk
  • Chat

Company

  • About
  • Pricing
  • Contact
  • Suggest a method

Entries are compiled from published sources for reference. Verifying the accuracy and suitability of any information for your own use remains your responsibility.

© 2026 ScholarGate · A research-method reference library
  • Privacy
  • Cookies
  • Terms
  • Delete account