I have a system and I want to find which properties hold for this system

$$ y(t) = \left.\begin{cases} x(t) & t \geq 1 \\ \\ 0 & -1 < t < 1 \\ \\ -x(t) & t \leq -1 \end{cases} \right\} \\ \\ $$

is the system linear, time invariant, causal , invertible, stable?

My problem is that I don't know how to work, because the function is piecewise defined.If I try to prove for example that each piece of function is linear then the whole function is linear ? With this methodology I found that the system is linear, causal and stable.Am I right?

  • \$\begingroup\$ Your piecewise function has conflicting limits, the first condition and the last condition both include the point t = 1. \$\endgroup\$ – KillaKem Apr 5 '14 at 18:13

The system is linear, causal, stable, but not time-invariant.

It is linear because from the definition if \$y_1(t)\$ is the response to input \$x_1(t)\$, and \$y_2(t)\$ is the response to input \$x_2(t)\$, then the response to the input signal \$ax_1(t)+bx_2(t)\$ is given by


It is causal because in order to compute the current output signal, not future values of the input signal are necessary, i.e. to compute \$y(t_0)\$ we only need to know \$x(t_0)\$ (and not even its past values, i.e. the system has no memory).

The system is stable in the BIBO (bounded-input bounded-output sense), because any bounded input signal \$|x(t)|<K\$ produces a bounded output signal \$|y(t)|<L\$ (in our case \$K=L\$).

The system is time-variant because the response to a shifted version of the input signal is not equal to the shifted output signal, i.e. if \$y(t)\$ is the response to \$x(t)\$, then \$y(t-t_0)\$ is generally not the response to \$x(t-t_0)\$. You can see this by noting that the output is always zero for \$-1<t<1\$, no matter how the input signal is shifted. If the system were time-invariant, then also the zero portion of the output would need to be shifted with the input signal.

  • \$\begingroup\$ A system is not causal if it is non-zero for negative values of t, so this system will only be causal in the trivial case of \$x(t) = 0\$, and if it is not causal then there will be no stability to talk of because the system will not even be realisable. \$\endgroup\$ – KillaKem Apr 5 '14 at 18:49
  • \$\begingroup\$ Please read my comment to your answer. I'm afraid your statements do not make much sense. Causality of a system does obviously not depend on its input signal, it is a property of the system. And of course there are non-causal stable systems. Please read up on some systems theory. \$\endgroup\$ – Matt L. Apr 5 '14 at 18:55
  • \$\begingroup\$ What makes you think \$x(t)\$ is an input? \$\endgroup\$ – KillaKem Apr 5 '14 at 18:58
  • \$\begingroup\$ Well, I interpret the system description as an input output relation with x(t) the input signal, and y(t) the output signal. What is your interpretation then? \$\endgroup\$ – Matt L. Apr 5 '14 at 19:01
  • \$\begingroup\$ I take \$y(t)\$ as being the impulse response and \$x(t)\$ as some unknown function of t. \$\endgroup\$ – KillaKem Apr 5 '14 at 19:07

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.