(6 intermediate revisions by one other user not shown)
Line 1: Line 1:
 
We are given the input to an LTI system along with the system's impulse response and told to find the output y(t).  Since the input and impulse response are given, we simply use convolution on x(t) and h(t) to find the system's output.
 
We are given the input to an LTI system along with the system's impulse response and told to find the output y(t).  Since the input and impulse response are given, we simply use convolution on x(t) and h(t) to find the system's output.
  
<math>y(t) = h(t) * x(t) = \int_{-\infty}^\infty h(\tau)x(t - \tau)d(\tau)</math>
+
<math>y(t) = h(t) * x(t) = \int_{-\infty}^\infty h(\tau)x(t-\tau)d\tau</math>
 +
 
  
 
Plugging in the given x(t) and h(t) values results in:
 
Plugging in the given x(t) and h(t) values results in:
  
<math>y(t) = \int_{-\infty}^\infty e^{-\tau}u(\tau)u(t - \tau - 1)d(\tau)</math>
+
<math>
 +
\begin{align}
 +
y(t) & = \int_{-\infty}^\infty e^{-\tau}u(\tau)u(t-\tau-1)d\tau \\
 +
      & = \int_0^\infty e^{-\tau}u(t-\tau-1)d\tau \\
 +
      & = \int_0^{t-1} e^{-\tau}d\tau  \\
 +
      & = 1-e^{-(t-1)}\, \mbox{ for } t > 1
 +
\end{align}
 +
</math>
  
<math>= \int_0^\infty e^{-\tau}u(t - \tau - 1)d(\tau)</math>
 
  
<math>= \int_0^{t-1} e^{-\tau}d(\tau) = 1 - e^{-(t - 1)}</math> for t > 1
+
Since x(t) = 0 when t < 1:
  
 +
<math>y(t) = 0\, \mbox{ for } t < 1</math>
  
Since x(t) = 0 when t < 1:
+
 +
<math>\therefore y(t) =  
 +
\begin{cases}
 +
  1-e^{-(t-1)},  & \mbox{if }t\mbox{ is} > 1 \\
 +
  0, & \mbox{if }t\mbox{ is} < 1  
 +
\end{cases}</math>
 +
 
 +
==Alternative Solutions==
 +
[[Problem 5 - Alternate Solution_Old Kiwi]]
  
<math>y(t) = 0\,</math> for t < 1
+
[[Problem 5 - Graphical Solution_Old Kiwi]]

Latest revision as of 16:23, 3 July 2008

We are given the input to an LTI system along with the system's impulse response and told to find the output y(t). Since the input and impulse response are given, we simply use convolution on x(t) and h(t) to find the system's output.

$ y(t) = h(t) * x(t) = \int_{-\infty}^\infty h(\tau)x(t-\tau)d\tau $


Plugging in the given x(t) and h(t) values results in:

$ \begin{align} y(t) & = \int_{-\infty}^\infty e^{-\tau}u(\tau)u(t-\tau-1)d\tau \\ & = \int_0^\infty e^{-\tau}u(t-\tau-1)d\tau \\ & = \int_0^{t-1} e^{-\tau}d\tau \\ & = 1-e^{-(t-1)}\, \mbox{ for } t > 1 \end{align} $


Since x(t) = 0 when t < 1:

$ y(t) = 0\, \mbox{ for } t < 1 $


$ \therefore y(t) = \begin{cases} 1-e^{-(t-1)}, & \mbox{if }t\mbox{ is} > 1 \\ 0, & \mbox{if }t\mbox{ is} < 1 \end{cases} $

Alternative Solutions

Problem 5 - Alternate Solution_Old Kiwi

Problem 5 - Graphical Solution_Old Kiwi

Alumni Liaison

Ph.D. on Applied Mathematics in Aug 2007. Involved on applications of image super-resolution to electron microscopy

Francisco Blanco-Silva