Complex QR Decomposition Using Fast Plane Rotations for MIMO Applications

Author

Awasthi, Abhishek ; Guttal, Rohit ; Al-Dhahir, Naofal ; Balsara, Poras T.

Author_Institution

Xtendwave, Dallas, TX, USA

Volume

18

Issue

10

fYear

2014

fDate

Oct. 2014

Firstpage

1743

Lastpage

1746

Abstract

QR decomposition (QRD) is widely used in various engineering applications and its implementation has a significant impact on the system performance and complexity. This paper develops a low-complexity QRD algorithm based on fast plane rotations, which does not require square-root operations for decomposing a complex-valued matrix. Furthermore, an update-based implementation is presented where computations are performed incrementally as the data arrives sequentially in time to drastically reduce the overall latency and hardware resources. Practical results for QRD-based spatial correlation estimator are provided to demonstrate the effectiveness of our solution for multiple-input multiple-output (MIMO) systems with complex-valued signals.

Keywords

MIMO communication; matrix algebra; MIMO; QR decomposition; complex-valued matrix; fast plane rotations; multiple-input multiple-output systems; spatial correlation estimator; square-root operations; Complexity theory; Hardware; MIMO; Matrix decomposition; Memory management; Noise; Least-squares problems; fast plane rotations; square-root free QR-decomposition;

fLanguage

English

Journal_Title

Communications Letters, IEEE

Publisher

ieee

ISSN

1089-7798

Type

jour

DOI

10.1109/LCOMM.2014.2349995

Filename

6881609