B matrix results.(MIMO Transceiver Design via Majorization
Theory)
B.1 Generalized Triangular Decompositions
The nonlinear MIMO transceiver designs in Chapter 4 are based on
the generalized triangular decomposition (GTD) of the form H =
QR[P.sup.[dagger]], where . . .
A convex optimization theory.(MIMO Transceiver Design via
Majorization Theory)
In the last two decades, several fundamental and practical results
have been obtained in convex optimization theory [12, 13, 20]. The
engineering community not only has benefited from these recent . . .
5 Extensions and future lines of research.(MIMO Transceiver
Design via Majorization Theory)
This text has developed a unified framework, based on majorization
theory, for the optimal design of linear and nonlinear decision-feedback
MIMO transceivers in point-to-point MIMO communication . . .
4.8 Summary.(4 Nonlinear Decision Feedback MIMO
Transceivers)
This chapter has introduced the design of nonlinear DF MIMO
transceivers with full CSI based on majorization theory and the
generalized triangular decomposition (GTD) algorithm. Two . . .
4.7 A particular case: CDMA sequence design.(4 Nonlinear Decision
Feedback MIMO Transceivers)
Similar to MIMO transceiver design, the symbol synchronous CDMA
sequence design problem also has been studied intensively over the past
decade (e.g., [55, 127, 161, 163]). However, these two topics . . .
4.6 A dual form based on dirty paper coding.(4 Nonlinear Decision
Feedback MIMO Transceivers)
In the preceding sections, we have studied the nonlinear MIMO
transceiver design as a combination of a linear precoder with an
MMSE-DFE. In this section, we introduce an alternative . . .
4.5 Optimum transmitter with individual QoS constraints.(4
Nonlinear Decision Feedback MIMO Transceivers)
This section deals with the problem formulation in (4.14)
reproduced next for convenience:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]. (4.73)
The optimal DFE matrices W and B have . . .
4.4 Optimum transmitter with global measure of performance.(4
Nonlinear Decision Feedback MIMO Transceivers)
This section deals with the design of the optimum transmitter P
according to problem (4.13). While the linear MIMO transceiver design in
Chapter 3 relies on the additive majorization theory, the . . .
4.3 Optimum decision feedback receiver.(4 Nonlinear Decision
Feedback MIMO Transceivers)
It has been observed in (4.7) that the optimal feedback matrix must
be
B = U([W.sup.[dagger]]HP), (4.15)
where U(x) stands for keeping the strictly upper triangular entries
of the matrix while . . .
4.2 Problem formulation.(4 Nonlinear Decision Feedback MIMO
Transceivers)
We now introduce the general problem of optimizing the nonlinear DF
MIMO transceiver, i.e., the precoder P, the DFE feed-forward filter
matrix W, and the feedback filter matrix B at the receiver. . . .
4.1 System model.(4 Nonlinear Decision Feedback MIMO
Transceivers)
In this chapter, we consider the same communication system as
introduced in Section 3.1. The received sampled baseband signal is
y = HPx + n, (4.1)
where [MATHEMATICAL EXPRESSION NOT . . .
4 Nonlinear decision feedback MIMO transceivers.(MIMO Transceiver
Design via Majorization Theory)
The preceding chapter focused on the design of linear MIMO
transceivers, i.e., the combination of a linear precoder with a linear
equalizer. In this chapter, we shall introduce another paradigm of . . .
3.8 Summary.(3 Linear MIMO Transceivers)
This chapter has considered the design of linear MIMO transceivers
with perfect CSI under a very general framework based on majorization
theory. Two different problem formulations have been . . .
3.7 Extension to multicarrier systems.(3 Linear MIMO
Transceivers)
As mentioned in Section 3.1, multicarrier systems (and some other
systems) may be more conveniently modeled as a communication through a
set of parallel and non-interfering MIMO channels:
. . .
3.6 Optimum linear transmitter with global measure of
performance: arbitrary cost functions.(3 Linear MIMO
Transceivers)
Section 3.4 dealt with the problem formulation with a global
measure of performance based on the family of Schur-concave and
Schur-convex functions. However, even though this family of . . .
3.5 Optimum linear transmitter with individual QoS constraints.(3
Linear MIMO Transceivers)
This section deals with the problem formulation with individual QoS
constraints and minimum transmit power as in (3.15).
The optimal receiver W has already been obtained in Section 3.3
(see . . .
3.4 Optimum linear transmitter with global measure of
performance: Schur-convex/concave cost functions.(3 Linear MIMO
Transceive
This section deals with problem formulations with a global measure
of performance, either as the minimization of a global cost function
subject to a power constraint as in (3.13) or the . . .
3.3 Optimum linear receiver.(3 Linear MIMO Transceivers)
The problem formulations in (3.13), (3.14), and (3.15) are very
difficult problems because they are nonconvex. To see this simply
consider the term [|[w.sup.[dagger].sub.i] H[p.sub.i] - 1|.sup.2] . . .
3.2 Problem formulation.(3 Linear MIMO Transceivers)
This section formulates the design of the linear MIMO transceiver
(matrices P and W) as a tradeoff between the power transmitted and the
performance achieved. First, using a global measure of . . .
3.1 System model.(3 Linear MIMO Transceivers)
The baseband signal model corresponding to a transmission through a
general MIMO communication channel with [n.sub.T] transmit and [n.sub.R]
receive dimensions is
y = Hs + n, (3.1)
where . . .
|
|