Dear Mohammad Sadegh Ayubirad ,

Look the link, maybe useful.

https://www.mathworks.com/matlabcentral/answers/482479-signal-decomposition-by-wavelet-synchrosqueezed-transform

Article Synchrosqueezed wavelet transforms: An empirical mode decomp...

Regards,

Shafagat

Dear Dr. Shafagat Mahmudova

Thanks, but my question is the same as the question you sent me from Mathworks Q&A. There's no answer there.

Dear Aparna Sathya Murthy

Thanks, I really appreciate it.

Analiza primero la naturaleza de la señal que está trabajando, para establecer la wavelet que más se correlaciona. Planifica las wavelets que presentan la mejor compresión empleando como parámetro la relación señal a ruido.

Dear Aparna Sathya Murthy

Thank you so much for your efforts, as you can see I don't know why "wsst" results are way different than those of the SWT function used in the previously mentioned paper. They're both synchrosqueezed wavelet transform and yet ...

The synchrosqueezed transform has reduced energy smearing when compared to the continuous wavelet transform. The input, x, must be a 1-D real-valued signal with at least four samples. wsst computes the synchrosqueezed transform using the analytic Morlet wavelet.

Dear Patrick Horn

Thanks for the answer, but it doesn't explain why MatLab's wsst results are not similar to the SWT results of the paper I'd attached.

I understand. I will look into this further and get back to you. I have found that MATLAB sometimes seems to have internal inconsistencies regarding certain variable equations. Like any program bugs do occur causing errors. The other option is perhaps your data is in error? I do not know enough of the computation at hand without checking the veracity of the data. I will definitely look into this and see if I can help you. Please have a good day Sir.