| Summary: | 碩士 === 國立中正大學 === 通訊工程研究所 === 99 === Lattice reduction (LR) has been proposed to improve the performance of
many multiple-input-multiple-output (MIMO) communications systems.
Among the LR algorithms, the LLL is the most practical algorithm to
date. More recently, LLL-deep has been proposed. It has been shown
that the LLL-deep is capable of finding shorter basis vectors than the
LLL, due to a more global swapping scheme (deep insertion).
This thesis focuses on the design of LR-aided MIMO communications
systems using the concept of LLL-deep. Practical issues including the
fixed complexity, early termination, and parallel processing have been
investigated, and two new algorithms using LLL-deep have been
proposed. The performances of the proposed algorithms under both
LR-aided precoding and LR-aided detection have been quantified by
computer simulations.
|
|---|
