0

The code I have in this question I just modified it to generate a non-stationary signals as mentioned below. I just want to know is it the correct way to generate a non-stationary signals?

Code

%% Time specifications:
   Fs = 8000;                       % samples per second
   dt = 1/Fs;                       % seconds per sample
   StopTime = 1;                    % seconds
   t = (0:dt:StopTime-dt);         % seconds

   x = (10)*cos(2*pi*3*(t-.2))...          
      + (20)*cos(2*pi*6*(t-.7))...
      + (20)*cos(2*pi*2*(t-.5));
Community
  • 1
  • 1
Amrmsmb
  • 1
  • 27
  • 104
  • 226
  • 2
    To me, "non-stationary" means that the statistics of the signal (given some statistical metric) is not constant through time. Your code creates a signal composed of three sine waves at three different frequencies. None of the properties of this signal change in time, so I would say that this is not a non-stationary signal. If you want a non-statinonary signal, you must first define what statistical property you want to vary thought time. – chipaudette Dec 29 '14 at 01:50

2 Answers2

1

No, its not the correct way because non stationary signal means that the properties of the signal does not remain constant, i.e the properties of the signal changes after some time but in your case at every point the signal is the sum of three cosine wave with same frequencies , amplitudes and phases.

Narendra
  • 19
  • 1
  • This does not provide an answer to the question. To critique or request clarification from an author, leave a comment below their post - you can always comment on your own posts, and once you have sufficient [reputation](http://stackoverflow.com/help/whats-reputation) you will be able to [comment on any post](http://stackoverflow.com/help/privileges/comment). – Anik Islam Abhi Feb 19 '15 at 08:49
  • Hello Anik, i have wrote a simple code for a non-stationary signal (Y), the signal is stationary only in small interval as from t=1:100 the frequency is 100 while from t=101:200, the frequency is 200 and from t=201:300 it is 300 but if we look at the signal Y from interval 1 to 300 the signal is non-stationary as the parameters of the signals are changing with time. – t=1:100; Y1=cos(100*t); t=101:200; Y2=cos(200*t); t=201:300; Y3=cos(300*t); Y=[Y1 Y2 Y3]; – Narendra Feb 20 '15 at 13:15
1

For different time intervals signal contains different frequency components in non stationary signal. The signal you have generated is stationary signal, as at any instance of time you have same frequency components. Speech signal as you record through microphone will have different components and is an example of non stationary signal. Another example of non stationary signal is ultrasonic A scan obtained in pulse echo testing. What Narendra generated can be called as nonstationary signal.

ravi gopal allamsetty
  • 11
  • 2