Abstract
This thesis deals with a wide range of topics within the research
area of advanced baseband receiver design for wireless communication
systems. In particular, the work focuses on signal processing
algorithms for receivers in multiple-input multiple-output (MIMO)
orthogonal frequency-division multiplexing (OFDM) systems, with a
particular emphasis on the 3rd Generation Partnership Project (3GPP)
Long Term Evolution (LTE) standard as a study case.
Signal processing in wireless receivers can be designed following
different strategies. On the one hand, one can use intuitive
argumentation to define the structure of the receiver with the hope
that the resulting heuristic architecture will exhibit the desired
behavior and performance. On the other hand, one can employ
analytical frameworks to pose the problem as the optimization of a
global objective function subject to certain constraints. This work
includes contributions based on both types of approaches.
Our work on analytical frameworks is mainly focused on tools from
variational Bayesian inference in probabilistic models and, more
specifically, the mean-field (MF) and belief propagation (BP)
methods. Within this context, one of our main contributions is the
derivation of a novel message-passing scheme combining the MF and BP
frameworks; the algorithm is derived from the stationary points of a
region-based free energy approximation, and is guaranteed to
converge if the underlying probabilistic model satisfies certain
conditions. Moreover, we apply the combined message-passing
algorithm to the probabilistic model of a MIMO-OFDM system; from the
general derivation of the messages in the model, several instances
of receiver structures with varying degrees of computational
complexity and performance are obtained. We also explore the
applicability of MF methods to the problem of estimation of sparse
signals.
Among the contributions within the area of heuristic approaches, we
highlight our study of iterative MIMO detection, interference
cancellation and decoding for LTE systems. A detailed study of
channel estimation algorithms for OFDM is also provided, including
both pilot-based and data-aided schemes.
area of advanced baseband receiver design for wireless communication
systems. In particular, the work focuses on signal processing
algorithms for receivers in multiple-input multiple-output (MIMO)
orthogonal frequency-division multiplexing (OFDM) systems, with a
particular emphasis on the 3rd Generation Partnership Project (3GPP)
Long Term Evolution (LTE) standard as a study case.
Signal processing in wireless receivers can be designed following
different strategies. On the one hand, one can use intuitive
argumentation to define the structure of the receiver with the hope
that the resulting heuristic architecture will exhibit the desired
behavior and performance. On the other hand, one can employ
analytical frameworks to pose the problem as the optimization of a
global objective function subject to certain constraints. This work
includes contributions based on both types of approaches.
Our work on analytical frameworks is mainly focused on tools from
variational Bayesian inference in probabilistic models and, more
specifically, the mean-field (MF) and belief propagation (BP)
methods. Within this context, one of our main contributions is the
derivation of a novel message-passing scheme combining the MF and BP
frameworks; the algorithm is derived from the stationary points of a
region-based free energy approximation, and is guaranteed to
converge if the underlying probabilistic model satisfies certain
conditions. Moreover, we apply the combined message-passing
algorithm to the probabilistic model of a MIMO-OFDM system; from the
general derivation of the messages in the model, several instances
of receiver structures with varying degrees of computational
complexity and performance are obtained. We also explore the
applicability of MF methods to the problem of estimation of sparse
signals.
Among the contributions within the area of heuristic approaches, we
highlight our study of iterative MIMO detection, interference
cancellation and decoding for LTE systems. A detailed study of
channel estimation algorithms for OFDM is also provided, including
both pilot-based and data-aided schemes.
| Original language | English |
|---|---|
| Place of Publication | Aalborg |
| Publisher | |
| Print ISBNs | 978-87-92328-77-9 |
| Publication status | Published - 2011 |