Signal Analysis: Time Segments & Characteristics

Click For Summary
SUMMARY

The discussion focuses on analyzing a signal in terms of time segments to understand its characteristics and patterns. The user has attempted various methods including RMS, Wavelet Transform, and Fast Fourier Transform (FFT) but seeks clarity on which method provides the best insights. Recommendations include creating a spectrogram for frequency analysis and using the Lomb-Scargle Periodogram for identifying underlying frequency components. MATLAB code snippets for generating power and amplitude spectra are provided to assist in the analysis.

PREREQUISITES
  • Understanding of signal processing concepts such as RMS and FFT.
  • Familiarity with MATLAB for implementing signal analysis techniques.
  • Knowledge of frequency domain analysis and its applications.
  • Basic principles of spectrograms and periodograms.
NEXT STEPS
  • Learn how to create and interpret spectrograms using MATLAB.
  • Investigate the Lomb-Scargle Periodogram for analyzing unevenly sampled data.
  • Explore advanced Wavelet Transform techniques for time-frequency analysis.
  • Study the implications of moving average methods in signal smoothing and analysis.
USEFUL FOR

Signal processing engineers, data analysts, and researchers interested in time-frequency analysis and signal characterization techniques.

Hzaqa
Messages
15
Reaction score
0

Homework Statement


Hi ,
The problem is how to analysis this signal in terms of time segments to understnd the characteristics/pattern of this signal
146493896.png


Homework Equations



There is no relevant equtions

The Attempt at a Solution



I have tried to use :
RMS
Wavlet
FFT
But which one of them can give us a better understing of the system,

Thanks,
 
Physics news on Phys.org
Your problem statement is fairly unconstrained so--

For frequency analysis, consider making a spectrogram, a 3 dimensional graph (frequency vs amplitude vs time).

For amplitude analysis, experiment with various moving average (RMS) plots.
 
Presumably you have a data set that corresponds to the graph, and not just the graph alone? You might want to investigate the use of the Lomb-Scargle Periodogram method to tease out the underlying frequency components of the signal.
 
Thanks for the response,


Well, I have applied frequency domain


Matlab code:

m=fft(x)
n=length(x)
power=abs(m(1:floor(n/2))).^2
nyquist=1/2
freq=((1:n/2)/(n/2)*nyquist)*5000
figure
plot(freq,power)
xlabel('Frequency (Hz)')
ylabel('Power')
title('Power Spectrum of Signal')
grid on
977009973.png


amp=abs(m(1:floor(n/2)))
figure
plot(freq,amp)
xlabel('Frequency (Hz)')
ylabel('Amplitude')
title('Amplitude Spectrum of Signal')
grid on
975500764.png





But I am not quite sure which one of them is best/correct,any suggestion,
Thanks
 

Similar threads

  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 2 ·
Replies
2
Views
1K
  • · Replies 13 ·
Replies
13
Views
6K
  • · Replies 3 ·
Replies
3
Views
6K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 9 ·
Replies
9
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K