回复： How to find the frequency bands of the DWT for the db04 wavelet?

[shahjigar]

Respected sir,

I am doing project on EEG signal simulation, processing and analysis. I want to find frequency band of signal after extracting the detail and Approximate coefficient. I used here wavelet analysis(db04). I will be gratfull to you if you can help me out in this matter.

  Thanking you.

[ZJ.Gu · Accepted solution]

Hi,

You could use WA Wavelet Filter.vi to obtain the filter coefficients used in wavelet transform, then you could anlayze the frequency responses.

Thanks!

ZJ Gu

[shahjigar]

Hi,

I used WA discrete wavelet transfor & WA get coefficient of discrete wavelet transform. my question was after obtain these signals how to find frequency band in which these signals lies?

Thank you

Jigar

[Macpac]

Hi Jigar,

We can observe the power spectral density of the signal to check the frequency band. After a slight modification to our existing example, I posted the VI to show the frequency band here. Hope it can help:)

Mac

[hubc@lboro.ac.uk]

Hi macpac, I'm using an older version of LabVIEW (8.5 Developer kit) for the Advanced Processing toolkit. Is it possible for you to send a compatible  version of your "Get and Set Single Level Detail Coefficients.vi" or an image of the block diagram please?

[Macpac]

Hi Jigar,

I modified the code using LV8.2.1. I guess the most direct way to check the frequency band is to calculate the power spectral density of the target signal and plot it on the PSD graph using Time Series Analysis VIs. Do you agree with me?

Mac

[shahjigar]

Hi,

I think,

First of all Used discrete wavelet transform then used Get wavelet coefficient. if starting frequency band of neuroelectric waveform is 0-64 Hz then at level 1 we will get 0-32 Hz which gives approximation coefficience & other band is 32-64Hz which gives Detail coefficient of wavelet. Further at level 2 we have 4 frequency band 1->0-16 Hz i.e.approximation coefficience

                                                                                                                                                                                                2->16-32 Hz i.e. Detail coefficient

                                                                                                                                                                                                 3->32-48 Hz i.e. approximation coefficience

                                                                                                                                                                                                 4->48-64 Hz i.e. Detail coefficient.

Further at Level 3 we have 8 Frequency band ,use same logic,  ......So and So..... Do you agree with me?

Thank You.

[Macpac]

Yes, that's correct. But please notice that our discrete wavelet transform will always return only 1 approximation coefficients. That is, in your case, if the neuroelectric waveform is 0-64Hz and levels = 2, the discrete wavelet transform will return 3 frequency bands:

1 -> 0-16 Hz i.e. approximation coefficients

2 -> 16-32 Hz i.e. detail coefficients

3 -> 32-64 Hz i.e. detail coefficients

At level 3, we will have 4 frequency bands: 0-8Hz, 8-16Hz, 16-32Hz, 32-64Hz. If you want to get all the 8 frequency bands, you may want to use the arbitrary path analysis VI instead of the discrete wavelet transform VI. Please let me know if you have any questions.

Mac

[shahjigar]

Thank you.

[fisico]

Hi Macpac,

Marco here. I read your good article on the bispectrum and bicoherence. Woudl you mind if I sent you a couple of clarifying questions if you have some time?

thanks

Marco

marcoscipioni1@gmail.com