How to calculate the instantaneous frequency for one specfic signal?

Question

I have a signal which like this: I want to calculate this signal's instantaneous frequency.I have tried hht and so on which method I can find but can not get the result. So does somebody can give me some advice? If can proveide some python program which will be must useful for me.Thanks!

UPDATED: Thanks @OverLordGoldDragon I use this code in this answer:

import numpy as np
from ssqueezepy import ssq_cwt, extract_ridges
from ssqueezepy.visuals import plot, imshow
z = see OP's code; used np.random.seed(1000)
beta = 24
Tx, Wx, ssq_freqs, scales, *_ = ssq_cwt(z, ('gmw', {'beta': beta}), padtype='zero', fs=6)
ridge_idxs = extract_ridges(Tx, scales, penalty=20)
plot(ridge_idxs, color='k', linestyle='--', xlims=(0, len(z) - 1))
imshow(Tx, abs=1, yticks=ssq_freqs[::-1], ylabel="Frequencies [Hz]",
       title="abs(SSQ_CWT), wavelet=('gmw', {'beta': %s})" % beta)
amplitude = np.abs(Tx[ridge_idxs[:, 0], np.arange(len(z))])
frequencies = ssq_freqs[::-1][ridge_idxs[:, 0]]
plot(amplitude, ylims=(0, None), title="Amplitude vs time, SSQ ridge", show=1)
plot(frequencies, ylims=(0, None), ylabel="Frequencies [Hz]", 
     title="Frequency vs time, SSQ ridge", show=1)

But my result like this: Which seems disorder.

score 0 · Accepted Answer · answered Jun 16 '21 at 23:33

0

Synchrosqueezing with ssqueezepy.

from ssqueezepy import TestSignals, ssq_cwt
from ssqueezepy.visuals import plot, imshow
x = TestSignals(N=2048).lchirp()[0]
Tx, , ssq_freqs, * = ssq_cwt(x)
plot(x, title="linear chirp", show=1)
imshow(Tx, abs=1, title="abs(SSQ_CWT)", xlabel="time", ylabel="frequency",
       yticks=ssq_freqs[::-1])

answered Jun 16 '21 at 23:33

OverLordGoldDragon

8,912
5
23
74

Thanks for you help! Dose ssqueezepy has some docs? I'm a newer for signal processing. I want to know which is the result of signal's instantaneous frequency and how to use this. Thanks again! – user57947 Jun 17 '21 at 01:53
@user57947 Every library function is documented, e.g. help(ssq_cwt). There's lots of reference material in the README, including this overview, and examples. For choosing wavelet parameters, see here. – OverLordGoldDragon Jun 17 '21 at 02:13
@user57947 There's also learning resources ("DSP Fundamentals"), beginner friendly. – OverLordGoldDragon Jun 17 '21 at 02:18
Thanks,I think what I need may be in Tx, But I found Tx is a multi dimension array. So what dimension is the instantaneous frequency? – user57947 Jun 17 '21 at 03:35
@user57947 dim0 is frequency, dim1 is time. If you prefer a 1D array, can be done with ridge extraction - see this answer. – OverLordGoldDragon Jun 17 '21 at 03:59
I had use the code you answered in that question.But I got the result like the updated in my question.I think this may not correct. – user57947 Jun 17 '21 at 05:21
@user57947 That means parameters in ssq_cwt or extract_ridges need adjusting. But this is now a separate question, which you're free to ask - best with signal data included for download, as .npy or .csv (or paste as list here if reasonably long). – OverLordGoldDragon Jun 17 '21 at 05:31
Thanks! I had paste the first part(which is the limit max size) of data in here. – user57947 Jun 17 '21 at 05:52
@user57947 You should ask a new question, now about "instantaneous frequency vs time as 1D array". That data isn't really enough, try here, or share it fully with multiple pastebins. – OverLordGoldDragon Jun 17 '21 at 06:23
Thanks a lot! I think I had got it which as what you said change the parameters in ssq_cwt and extract_ridges.If I have some new questions I will ask you again. So please don't bored with me! – user57947 Jun 17 '21 at 06:52

How to calculate the instantaneous frequency for one specfic signal?

z = see OP's code; used np.random.seed(1000)

1 Answers1