Scribd
Upload
75 views

Empirical Mode Decomposition: and Hilbert-Huang Transform

1. Empirical Mode Decomposition (EMD) is a method for decomposing complicated data sets into Intrinsic Mode Functions (IMFs) through a sifting process. 2. The sifting process involves identifying local extrema and interpolating to create envelopes to extract IMFs. 3. Hilbert transforms are applied to the IMFs to determine instantaneous frequencies and construct a 3D energy-frequency-time distribution.

Uploaded by

Shaik Afroz
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
75 views

Empirical Mode Decomposition: and Hilbert-Huang Transform

1. Empirical Mode Decomposition (EMD) is a method for decomposing complicated data sets into Intrinsic Mode Functions (IMFs) through a sifting process. 2. The sifting process involves identifying local extrema and interpolating to create envelopes to extract IMFs. 3. Hilbert transforms are applied to the IMFs to determine instantaneous frequencies and construct a 3D energy-frequency-time distribution.

Uploaded by

Shaik Afroz
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 27

Empirical Mode Decomposition

and Hilbert-Huang Transform

Emine Can 2010

too short total data span

nonstationary

nonlinear process

Data Analysis
Fourier Spectral Analysis

Data Processing Methods Energy-frequency distributions =SpectrumFourier Transform of the data o Spectrogram Restrictions: * the system must be linear o Wavelet Analysis * the data must be strictly periodic or stationary Modifications of Fourier SA o Wigner-Ville Distribution (Heisenberg wavelet) o Evolutionary spectrum o Empirical Orthogonal Function Expansion (EOF) o Other methods
Empirical Mode Decomposition and Hilbert-Huang Transform

10.2010

Complicated Data Set

Empirical Mode Decomposition

Intrinsic Mode Functions

Instantaneous Frequency

(Energy-Frequency-Time)

Hilbert Transform

10.2010

Empirical Mode Decomposition and Hilbert-Huang Transform

Emperical Mode Decomposition (EMD) A method that any complicated data set can be decomposed into a finite and often small number of `intrinsic mode functions' that admit well-behaved Hilbert Transforms.
Intrinsic Mode Functions(IMF)

IMF is a function that satisfies two conditions: 1- In the whole data set, the number of extrema and the number of zero crossings must either equal or differ at most by one 2- At any point, the mean value of the envelope defined by the local maxima and the envelope defined by the local minima is zero
10.2010 Empirical Mode Decomposition and Hilbert-Huang Transform 4

The empirical mode decomposition method: the sifting process

10.2010

Empirical Mode Decomposition and Hilbert-Huang Transform

10.2010

Empirical Mode Decomposition and Hilbert-Huang Transform

The sifting process


Complicated Data Set x(t)

IMF 1; iteration 0 2 1 0 -1 -2 10 20 30 40 50 60 70 80 90 100 110 120

The sifting process


1. identify all upper extrema of x(t).

IMF 1; iteration 0 2 1 0 -1 -2 10 20 30 40 50 60 70 80 90 100 110 120

The sifting process


2. Interpolate the local maxima to form an upper envelope u(x).

IMF 1; iteration 0 2 1 0 -1 -2 10 20 30 40 50 60 70 80 90 100 110 120

The sifting process


3. identify all lower extrema of x(t).

IMF 1; iteration 0 2 1 0 -1 -2 10 20 30 40 50 60 70 80 90 100 110 120

The sifting process


4. Interpolate the local minima to form an lower envelope l(x).

IMF 1; iteration 0 2 1 0 -1 -2 10 20 30 40 50 60 70 80 90 100 110 120

The sifting process


5. Calculate the mean envelope: m(t)=[u(x)+l(x)]/2.

IMF 1; iteration 0 2 1 0 -1 -2 10 20 30 40 50 60 70 80 90 100 110 120

The sifting process


6. Extract the mean from the signal: h(t)=x(t)-m(t)
IMF 1; iteration 0 2 1 0 -1 -2 10 20 30 40 50 60 70 80 90 100 110 120

residue 1.5 1 0.5 0 -0.5 -1 -1.5 10 20 30 40 50 60 70 80 90 100 110 120

The sifting process


7. Check whether h(t) satisfies the IMF condition. YES: h(t) is an IMF, stop sifting. NO: let x(t)=h(t), keep sifting.
IMF 1; iteration 1 1.5 1 0.5 0 -0.5 -1 -1.5 10 20 30 40 50 60 70 80 90 100 110 120

residue 1.5 1 0.5 0 -0.5 -1 -1.5 10 20 30 40 50 60 70 80 90 100 110 120

The sifting process


IMF 1; iteration 1 1.5 1 0.5 0 -0.5 -1 -1.5 10 20 30 40 50 60 70 80 90 100 110 120

residue 1.5 1 0.5 0 -0.5 -1 -1.5 10 20 30 40 50 60 70 80 90 100 110 120

The sifting process


IMF 1; iteration 1 1.5 1 0.5 0 -0.5 -1 -1.5 10 20 30 40 50 60 70 80 90 100 110 120

residue 1.5 1 0.5 0 -0.5 -1 -1.5 10 20 30 40 50 60 70 80 90 100 110 120

The sifting process


IMF 1; iteration 1 1.5 1 0.5 0 -0.5 -1 -1.5 10 20 30 40 50 60 70 80 90 100 110 120

residue 1.5 1 0.5 0 -0.5 -1 -1.5 10 20 30 40 50 60 70 80 90 100 110 120

The sifting process


IMF 1; iteration 1 1.5 1 0.5 0 -0.5 -1 -1.5 10 20 30 40 50 60 70 80 90 100 110 120

residue 1.5 1 0.5 0 -0.5 -1 -1.5 10 20 30 40 50 60 70 80 90 100 110 120

The sifting process


IMF 1; iteration 1 1.5 1 0.5 0 -0.5 -1 -1.5 10 20 30 40 50 60 70 80 90 100 110 120

residue 1.5 1 0.5 0 -0.5 -1 -1.5 10 20 30 40 50 60 70 80 90 100 110 120

The sifting process


IMF 1; iteration 1 1.5 1 0.5 0 -0.5 -1 -1.5 10 20 30 40 50 60 70 80 90 100 110 120

residue 1.5 1 0.5 0 -0.5 -1 -1.5 10 20 30 40 50 60 70 80 90 100 110 120

10.2010

Empirical Mode Decomposition and Hilbert-Huang Transform

21

The signal is composed of 1. a high frequency triangular waveform whose amplitude is slowly (linearly) growing. 2. a middle frequency sine wave whose amplitude is quickly (linearly) decaying 3. a low frequency triangular waveform

10.2010

Empirical Mode Decomposition and Hilbert-Huang Transform

22

The sifting process


Stop criterion

A criterion for the sifting process to stop: Standard deviation, SD, computed from the two consecutive sifting results is in limited size.

:residue after the kth iteration of the 1st IMF A typical value for SD can be set between 0.2 and 0.3.

10.2010

Empirical Mode Decomposition and Hilbert-Huang Transform

23

Hilbert Transform
* Analytic Signal:

Instantaneous Frequency:

10.2010

Empirical Mode Decomposition and Hilbert-Huang Transform

24

Advantages
*Adaptive,highly efficient,applicable to nonlinear and non-stationary processes.
Traditional methods - Not appropriate for nonlinear & nonstationary signals. - Predefined basis and/or system model. - Distorted information extracted. - Full theoretical basis. EMD - Adequate for both nonlinear & nonstationary. - Adaptive data driven basis. - Preserves physical meaning. - Sharper spectrum - Lack of theoretical analysis.

10.2010

Empirical Mode Decomposition and Hilbert-Huang Transform

25

Applications of EMD
nonlinear wave evolution, climate cycles, earthquake engineering, submarine design, structural damage detection, satellite data analysis, turbulence flow, blood pressure variations and heart arrhythmia, non-destructive testing, structural health monitoring, signal enhancement, economic data analysis, investigation of brain rythms Denoising
Empirical Mode Decomposition and Hilbert-Huang Transform

10.2010

26

References

The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis Huang et al., The Royal Society, 4 November 1996. Rilling Gabriel, Flandrin Patrick , Goncalv`es Paulo, On Empirical Mode Decomposition and Its Algorithms Stephen McLaughlin and Yannis Kopsinis.ppt Empirical Mode Decomposition: A novel algorithm for analyzing multicomponent signals Institute of Digital Communications (IDCOM) Hilbert-Huang Transform(HHT).ppt Yu-Hao Chen, ID:R98943021, 2010/05/07

10.2010

Empirical Mode Decomposition and Hilbert-Huang Transform

27

You might also like