SEARCH
0-9 A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
Prev | Current Page 748 | Next

Joseph Boccuzzi

"Signal Processing for Wireless Communications"

8.9, assuming SNR2 SNR1.
We need to consider a system with memory of length M. Let??™s define the system as follows: Y is
to be determined by M independent samples.
(8.50) Y  G(x1, x2,c, xM)
SNR2 
PS
s2*
SNR1 
PS
s2
r 
N
N*
<
3
`
vth
fN(n)dn
3
`
vth
fN(n) # w(n)dn
s2(P^
0)  s2(P^
0*)
r 
N
N*
<
3
?
vth
f(v) dv
3
?
vth
f(v) # w(v) dv
468 CHAPTER EIGHT
Eb/No
(dB)
BER
(Eb/No)2
BER2
BER1
(Eb/No)1
FIGURE 8.9 IS functional application.
Then the weight function becomes
(8.51)
where is an M-dimensional PDF of the noise source. Assuming independent noise samples,
we arrive at the product of the individual noise statistics.
(8.52)
The memory in the system has a negative impact on the system improvement. The reader can visit
the applicable references in the end of this chapter to see various examples, in order to provide numerical
results (see [3] and [8]). Some noteworthy observations are, as the memory increases, the
improvement of IS over MC decreases. Also the optimum value of 1   varies as the length of the
memory changes.
Let us build a simple IS system and compare the error rate estimates. The noise samples are scaled
to produce a larger variance and more frequent error events.


Pages:
736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760
hotel jelenia góra Russian bride Free English grammar and study guid powiekszenia wielkoformatowe counter strike 1.6