![\begin{displaymath}\left(\frac{1}{2}\right)^n u[-n] =
\left(\frac{1}{2}\right)^n u[-n-1] + \delta(n)
\end{displaymath}](img1.gif) 
 
 
 
 .
.
 and 
2 < z < 3.  Thus, there are two possible two-sided sequences.
and 
2 < z < 3.  Thus, there are two possible two-sided sequences.
 is causal, but the poles at 2 and 3 are
non-causal.
is causal, but the poles at 2 and 3 are
non-causal.
 
 
 .
However the pole of
X(z) which limits the ROC to be less than 1 is cancelled by the zero of
H(z) (the u[-n-1] part of x[n] is cancelled), so the ROC of X(z) which
remains is
.
However the pole of
X(z) which limits the ROC to be less than 1 is cancelled by the zero of
H(z) (the u[-n-1] part of x[n] is cancelled), so the ROC of X(z) which
remains is 
 .
The ROC of Y(z) is the region
which satifies the remaining constraint on X(z) (|z| >
.
The ROC of Y(z) is the region
which satifies the remaining constraint on X(z) (|z| > 
 )
and the constraint on H(z) (|z| > 1).  Thus, Y(z) converges on |z| >
1.
)
and the constraint on H(z) (|z| > 1).  Thus, Y(z) converges on |z| >
1.
 
![\begin{displaymath}y[n] = -\frac{1}{3}\left(\frac{1}{2}\right)^n u[n] + \frac{1}{3}(-1)^n
u[n]\end{displaymath}](img13.gif) 
 
 .
.
 
 .
.
 
![\begin{displaymath}x[n] = -(2)^nv[-n-1]+\frac{1}{2}(2)^{n-1}u[-n]\end{displaymath}](img18.gif) 
 
![\begin{displaymath}h[n] = 2\left(\frac{1}{2}\right)^nu[n] - \left(-\frac{1}{4}\right)^nu[n]\end{displaymath}](img19.gif) 
 .
The ROC includes the unit circle, so the Fourier transfrom
converges.
.
The ROC includes the unit circle, so the Fourier transfrom
converges.
 ,
or
,
or 
 .
The ROC includes the unit circle, so the Fourier transfrom converges.
.
The ROC includes the unit circle, so the Fourier transfrom converges.
 
 
 
 .
.
 
 .
The ROC of
Y(z) is the intersection of the ROC of H(z) and the ROC of the left over
part of X(z).  Since the ROC of Y(z) is
.
The ROC of
Y(z) is the intersection of the ROC of H(z) and the ROC of the left over
part of X(z).  Since the ROC of Y(z) is 
 ,
the ROC of
H(z) must be
,
the ROC of
H(z) must be 
 .
The pole-zero plot of H(z) will have a
pole at
.
The pole-zero plot of H(z) will have a
pole at 
 and a zero at 2.
and a zero at 2.
 
![\begin{displaymath}h[n] = \left(\frac{2}{3}\right)^n u[n] -
2\left(\frac{2}{3}\right)^{n-1}u[n-1] =
\left(\frac{2}{3}\right)^n(u[n]-3u[n-1])\end{displaymath}](img31.gif) 
 
 
![\begin{displaymath}y[n] - \frac{2}{3}y[n-1] = x[n]-2x[n-1]\end{displaymath}](img34.gif) 
 .
.
 
 .
.
![\begin{displaymath}h[n] = 3 \left(-\frac{1}{3}\right)^n i[n]\end{displaymath}](img37.gif) 
![\begin{displaymath}y[n]=\sum_{n=-\infty}^{\infty}h[k]x[n-k]=
\sum_{n=-\infty}^{\infty}\left(3{\left( -\frac{1}{3}\right)}^k u[k]\right)u[n-k]
\end{displaymath}](img38.gif) 
![\begin{displaymath}= \sum_{k=0}^{n}3\left(-\frac{1}{3}\right)^k
= \frac{9}{4}\left(1-{\left(-\frac{1}{3}\right)}^{n+1}\right)u[n]
\end{displaymath}](img39.gif) 
 
![\begin{displaymath}y[n] = \frac{3}{4}\left(-\frac{1}{3}\right)^nu[n] + \frac{9}{4}u[n]\end{displaymath}](img41.gif) 
![\begin{displaymath}= \frac{9}{4}\left(1-{\left(-\frac{1}{3}\right)}^{n+1}\right)u[n]
\end{displaymath}](img42.gif) 
 
 
 
 
 .
.
 
 .
.
 
 .
This will make sure ROC of Y(z) is
.
This will make sure ROC of Y(z) is 
 (note that the zero of H(z) at z=2 cancels the pole of
X(z) at z=2, so the ROC of Y(z) is not limited to the region |z| < 2).
(note that the zero of H(z) at z=2 cancels the pole of
X(z) at z=2, so the ROC of Y(z) is not limited to the region |z| < 2).
 
![\begin{displaymath}h[n] =
\left(\frac{3}{4}\right)^nu[n]-2\left(\frac{3}{4}\right)^{n-1}u[n-1]\end{displaymath}](img51.gif) 
 
 
![\begin{displaymath}y[n] - \frac{3}{4}y[n-1] = x[n]-2x[n-1]\end{displaymath}](img53.gif)