How to do PSD measuremnet for the given signal

2 views (last 30 days)
Sarfaraz Ahmed
Sarfaraz Ahmed on 5 Jul 2021
Edited: Sarfaraz Ahmed on 22 Jul 2021
Hi, I am doing PSD (ms2) analysis for the frequency spectrum attached in the figure. I am able to find PSD but the results are not in PSD(ms2).
Here is the code:
[pxx,f] = pwelch(rwave_decimate_1,[],[],[],fs/N);
%the mean
mean = 0;
for i=1:length(pxx)
mean = mean + pxx(i);
M = mean/length(pxx);
%the Varaince
var = 0;
for j=1:length(pxx)
var = var + (pxx(j)-M)^2;
V = var/length(pxx);
It plots the PSD as in the attached figure. however, I want PSD in (ms2 or s2) on yaxis as in the figure 2, to make proper analysis on LF/HF region. Also, how can I highlight the LF and HF region with some shadded areas as in the figure2 ?
:- ms2 = mean-variance square
it would be appreciated if someone give solution.
Thanks in the advacnce.
Sarfaraz Ahmed
Sarfaraz Ahmed on 6 Jul 2021
Thank you for correction.
I tried like the way you mentioned, however, it gives only 1 mean and 1 variance value.
What I want is that PSD(s^2/hz) values on y axis, not in by default unit. I mean i want something like in the figure2.
In my case, when I plot(f,pxx) -> it does not plot PSD in s^2/hz, instead it just plot something over frequency.
so, can you please assist how can we plot PSD in s^2/hz ?
Thank you.

Sign in to comment.

Answers (1)

Arthi Sathyamurthi
Arthi Sathyamurthi on 19 Jul 2021
Hello Sarfaraz,
The units of the PSD estimate from the pwelch function are in squared magnitude units of the time series data per unit frequency. Thus, if the input data, rwave_decimate is in ms and the sample rate given as fs/N is in Hz, the PSD estimate which in your case pxx, ms^2/Hz.
Further, to highlight the LF and HF region with some shaded areas use the area function in MATLAB. You can also have a look at the answered question here for an example.
Sarfaraz Ahmed
Sarfaraz Ahmed on 22 Jul 2021
Thank you.
Actually, its not the case. rwave_decimate is the signal after passing extrated r_waves signal to low pass filter. I don't think its already in the ms/Hz.

Sign in to comment.


Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by