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 441 | Next

Joseph Boccuzzi

"Signal Processing for Wireless Communications"

116)
where d is the desired signal. The solution to the above criterion leads to the following equation for
the antenna weights:
(5.117)
where the received signal??™s covariance matrix and cross-correlation vector are defined and given as below.
(5.118)
(5.119)
In practice we have to estimate these parameters using the received signal since the channel is time
varying. We can replace the expectation operator by a sample mean estimator, which gives us the following
estimate of the covariance matrix.
(5.120)
And the following estimate of the cross-correlation vector
(5.121)
We have used N to denote the number of time samples used in the estimate calculation. The use of
these estimates is sometimes referred to as Sample Matrix Inversion (SMI) or Direct Matrix Inversion
(DMI) [59, 64??“66]. Hence the estimated MMSE array weights are given as
(5.122) w^
MMSE  R ^
1
xx # r^
xd
r^
xd 
1
NaN
i1
x(t  ti) # d*(t  ti)
R ^
xx 
1
NaN
i1
x(t  ti) # x*(t  ti)
rxd  E5x # d*6 Rxx  E5x # x*6
wMMSE  R1
xx # rxd
min
w
Ee gw* # x  d g2 f
wMMSE
276 CHAPTER FIVE
..
.
Weight
Calculation
X
X
X
Desired Signal, d
+
Decision
Device
x1
x2
xM
w1
w2
wM
y
FIGURE 5.65 Adaptive antenna array block diagram.


Pages:
429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453
hotel jelenia góra Russian bride Free English grammar and study guid powiekszenia wielkoformatowe counter strike 1.6