51)
where is the cross-correlation vector between the received signal and the reference (desired) signal,
denoted as d.
In the sections that follow, we vary the channel conditions gradually creating a more and more
complex covariance matrix, along the way we provide insight into the components that make up the
covariance matrix and correlation vector. We will emphasize the DMI method to calculate the STE
weights. More specifically we will utilize the ESD of the covariance matrix to provide insight into the
STE weights. This will be accomplished using the following spatio-temporal signal processing
receiver, shown with M 2 antennas (see Fig. 6.11).
rxd
wMMSE R1 # rxd
s2n
R H s2n
I
R cRxx Rxy
R*
xy Ryy d
R Eecx(k)
y(k) d# cx(k)
y(k) d* f
y(k) D y(k)
y(k T/2)
(
y(k (K 1)T/2)T x(k) D x(k)
x(k T/2)
(
x(k (K 1)T/2)T
RECEIVER DIGITAL SIGNAL PROCESSING 305
Estimated
Space-Time
Matrix, R
Space-Time Weight
Calculation &
Compensation
Eigen Spectra
Manager (ESM)
Cross-Correlation
Vector
Array
Output
FIGURE 6.11 Unified eigenbased, spatio-temporal signal processing block
diagram.
The proposed receiver continually estimates the eigenspectra of the received signal and uses a
priori information (obtained through extensive simulation and field trial campaigns) to adjust the
STE weights.
Pages:
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509