[Nonfractal-commits] how to define the scale range

You, Wonsang WYou at childrensnational.org
Tue Oct 17 13:19:22 PDT 2017


Hello Huibin,

1. Let f_N be the Nyquist frequency. The wavelet scale j is related to the frequency range [f_N/2^{j+1}, f_N/2^j]. So, if the sample rate = 1 second and you set up the option range=[1, 5], it would correspond to the frequency range [0.015625, 0.5] Hz.

2. The signal length should be a power of 2 in DWT. However, the input signal to bfn_mfin_ml is not necessary to have the length of power of 2 since it will automatically truncate the signal; for example, the signal with 100 will be truncated as 2^6=64.

I wish it is helpful for you. Thank you.

Best Regards,
Wonsang


From: nonfractal-commits-bounces at www.nitrc.org [mailto:nonfractal-commits-bounces at www.nitrc.org] On Behalf Of Hui-bin Jia
Sent: Monday, October 16, 2017 10:33 AM
To: nonfractal-commits <nonfractal-commits at www.nitrc.org>
Subject: [Nonfractal-commits] how to define the scale range

Dear Wonsang,
      I have only limited knowledge about DWT. In the function bfn_mfin_ml we need to define the scale range. I think this option is used to define the frequency range in Hurst estimation.
Usually, the frequency range in fMRI analysis is defined to be [0.01 0.1] Hz. Could you give me some tips about how to define the scale range according to the frequency range?

In some papers, the authors argue that the signal length should be a power of 2 in order to obtain DWT-based Hurst exponent. I'm asking whether the signal length should be a power of 2 in function bfn_mfin_ml?

Thank you!

Best Regards,
Huibin Jia

,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
Confidentiality Notice: This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.nitrc.org/pipermail/nonfractal-commits/attachments/20171017/88ffde99/attachment.html>


More information about the nonfractal-commits mailing list