Reconfigurable FPGA-Based Unit for Singular Value Decomposition of Large m x n Matrices
- 1 November 2011
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 23256532,p. 345-350
Abstract
Singular value decomposition (SVD) allows the factorization of real or complex matrices providing quantitative information with fewer dimensions along which data points exhibit more variation. These days SVD computation is being used in numerous applications, and because of its importance, different approaches for SVD hardware computation have been proposed, however, their application is limited by the inherent SVD calculation complexity making it possible to analyze up to 8 × 8 matrices until now, complying certain constrains like symmetry. This paper presents a generic and novel FPGA-based hardware architecture for SVD computation on large m × n matrices utilizing Hestenes approach and one-side Jacobi rotations. Four different study cases (2 × 2, 8 × 7, 16 × 32, and 32 × 127 matrices) validate the performance of the FPGA-based computation unit reaching a maximum estimation error of 3.3718% in the SVD estimation of a large matrix.Keywords
This publication has 14 references indexed in Scilit:
- Constellation Precoded Multiple BeamformingIEEE Transactions on Communications, 2011
- Adaptive Denoising by Singular Value DecompositionIEEE Signal Processing Letters, 2011
- EXIT Chart-Based Power Allocation for Iterative Frequency Domain MIMO DetectorIEEE Transactions on Signal Processing, 2011
- Singular Value Decomposition Hardware for MIMO: State of the Art and Custom Design2010
- FPGA based singular value decomposition for image processing applications2008
- Hardware-efficient steering matrix computation architecture for MIMO communication systems2008
- A Novel Scheme for the Parallel Computation of SVDsLecture Notes in Computer Science, 2006
- An FPGA-Based Singular Value Decomposition Processor2006
- A survey of CORDIC algorithms for FPGA based computers1998
- Inversion of Matrices by Biorthogonalization and Related ResultsJournal of the Society for Industrial and Applied Mathematics, 1958