2
$\begingroup$

$$y(t) = \int_{0}^{\infty} x(\tau)h(t-\tau) \ \mathrm{d}\tau$$ where $h(t) = u(t)-u(t-1)$.

I have some problem with understanding whether system LTI or LTV just looking the impulse response of the system. Is there any short-cut to identify the system time-variance looking the impulse function?

$\endgroup$
  • $\begingroup$ LTV systems do not have one impulse response, they have infinitely many. $\endgroup$ – Rodrigo de Azevedo Jan 12 '18 at 10:38
1
$\begingroup$

$h(t)$ can be rewritten as $\mathrm{rect}(t-0.5)$. So basically your output signal can be calculated as:

$$y(t) = \int_{0}^{\infty} x(\tau)\mathrm{rect}(t-0.5-\tau) \ \mathrm{d}\tau$$

Note that if $t<0$, then $t-0.5-\tau<-0.5$, as $\tau\in(0,\infty$). Ergo,

$$|t-0.5-\tau|>|0.5|$$

This means that the $\mathrm{rect}()$ function will be zero if $t$ takes negative values. Therefore, we get that $y(t)=0 \ \forall t<0$.

For $t>0$, all we have to do is convolve the rectangular window with $x(t)u(t)$, as the integral begins at $0$ and not $-\infty$.

If we call the window $w(t)$, we can express $y(t)$ as:

$$y(t)= \left\{ \begin{array}{ll} [x(t)u(t)]*w(t) & \mbox{if } t > 0 \\ 0 & \mbox{if } t < 0 \end{array} \right.$$

Now, could you determine whether the system is time-invariant or not?

Hint: try appling a delay to the input signal, and see if its output is equal to a delayed version of the output itself.

$\endgroup$
  • $\begingroup$ My comment was only valid for a discrete-time system; either I misread or the notation was changed, so now it's continuous-time and the claim $u(t)-u(t-1)=\delta(t)$ is not correct anymore (if $t$ is a continuous variable). $\endgroup$ – Matt L. Jan 11 '18 at 15:24
  • $\begingroup$ @MattL. Thanks for that, I messed up big time. I've already corrected it, I believe now the answer makes sense. $\endgroup$ – Tendero Jan 11 '18 at 15:51
2
$\begingroup$

If a system is described by an impulse response $h(t)$, then by definition it is an LTI system. Because impulse response of the form $h(t) = \mathcal{T} \{ \delta(t) \} $ only exists for LTI systems.

$\endgroup$
  • $\begingroup$ text book says it is time varying i am also thinking like you $\endgroup$ – Nail Tosun Jan 11 '18 at 12:04
  • $\begingroup$ @NailTosun What textbook? $\endgroup$ – Tendero Jan 11 '18 at 12:23
  • $\begingroup$ I edited the question $\endgroup$ – Nail Tosun Jan 11 '18 at 13:46

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.