0
$\begingroup$

I've been attempting to plot the following function using MATLAB: $$ x(k)=\sum_{n=11}^{50} \sqrt{n} \sin (2n\pi k) +\sum_{n=1}^{40}\sqrt[3]{n} \sin (3n\pi k) $$

Note that $k$ is a continuous variable.

The problem I am facing with MATLAB is when I plot it, I get a signal which has a fundamental period of 2 seconds, and not 1.5 seconds.

Have a look below: Plot using MATLAB plot function

I've tried a higher sampling rate: At a higher sampling rate

So, what is going on wrong? Why isn't the fundamental period 1.5Hz?

$\endgroup$
  • $\begingroup$ Assuming your horizontal axis is in seconds, you have 3 full cycles in 2 seconds, which gives a fundamental period of 3/2 Hz = 1.5 Hz, as you were expecting. $\endgroup$ – Jazzmaniac Mar 12 '16 at 23:34
  • $\begingroup$ @MattL. Thanks, Matt, but $x(k)$ is defined as a continuous periodic function. You could replace k by t. $\endgroup$ – Zushauque Mar 13 '16 at 6:37
  • $\begingroup$ @ Jazzmaniac Thanks for pointing that out, but... how do you identify the three cycles visually? I can't see them, or see a form of $x(n+N)$ which implies periodicity. $\endgroup$ – Zushauque Mar 13 '16 at 6:38
0
$\begingroup$

With $k$ a continuous variable, the frequencies of the sinusoids in the first sum are $f_n=n$, i.e. they are integers. The frequencies in the second sum are integer multiples of $\frac32$. The highest fundamental frequency of which all those frequencies are integer multiples is $f_0=\frac12$. The frequencies in the first sum are given by $2mf_0$ (with integer $m$), and the frequencies in the second sum are $3mf_0$. With a fundamental frequency of $f_0=\frac12$, the fundamental period is $T=2$, as observed by you.

$\endgroup$
  • $\begingroup$ Thanks a lot, Matt! That solved it. But could you explain how did you analytically calculate it? $\endgroup$ – Zushauque Mar 13 '16 at 12:38
  • $\begingroup$ @Zushauque: Well, the frequencies in your signal are given, and you just need to find a fundamental frequency such that all frequencies in the signal can be represented as integer multiples of that fundamental frequency. $\endgroup$ – Matt L. Mar 13 '16 at 18:27
  • $\begingroup$ Oh, did that. Thanks for telling me about that. By the way, is there is a systematic way in which I would calculate the fundamental frequency from the arguments of a series of the terms? I saw something about Greatest Common Divisor and Least Common Multiple, 1/2 is the LCM for this example, but it isn't really an integer which is kinda odd for the LCM? $\endgroup$ – Zushauque Mar 15 '16 at 19:25
  • $\begingroup$ @Zushauque: It's more like a GCD. Have a look at this site. $\endgroup$ – Matt L. Mar 15 '16 at 19:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.