Complexity and search space reduction in cyclic-by-row PEVD algorithms

Fraser K. Coutts; Jamie Corr; Keith Thompson; Stephan Weiss; Ian K. Proudler; John G.  McWhirter

Complexity and search space reduction in cyclic-by-row PEVD algorithms

Fraser K. Coutts, Jamie Corr, Keith Thompson, Stephan Weiss, Ian K. Proudler, John G. McWhirter

Electronic And Electrical Engineering

Research output: Contribution to conference › Paper › peer-review

84 Downloads (Pure)

Abstract

In recent years, several algorithms for the iterative calculation of a polynomial matrix eigenvalue decomposition (PEVD) have been introduced. The PEVD is a generalisation of the ordinary EVD and uses paraunitary operations to diagonalise a parahermitian matrix. This paper addresses potential computational savings that can be applied to existing cyclic-by-row approaches for the PEVD. These savings are found during the search and rotation stages, and do not significantly impact on algorithm accuracy. We demonstrate that with the proposed techniques, computations can be significantly reduced. The benefits of this are important for a number of broadband multichannel problems.

Original language	English
Number of pages	5
Publication status	Accepted/In press - 19 Jul 2016
Event	50th Asilomar Conference on Signals, Systems and Computers - Asilomar Conference Ground, Pacific Grove, CA, United States Duration: 6 Nov 2016 → 9 Nov 2016 http://www.asilomarsscconf.org/

Conference

Conference	50th Asilomar Conference on Signals, Systems and Computers
Abbreviated title	Asilomar 2016
Country	United States
City	Pacific Grove, CA
Period	6/11/16 → 9/11/16
Internet address	http://www.asilomarsscconf.org/

Keywords

polynomial matrix eigenvalue decomposition
PEVD
space-time covariance matrices
algorithmic cost reductions
SMDCbR
sequential matrix diagonalisation

Access to Document

Coutts-etal-ACSSC-2016-Complexity-and-search-space-reduction-in-cyclic-by-row-PEVD-algorithmsAccepted author manuscript, 147 KB

http://www.asilomarsscconf.org/

1 Finished

Signal Processing Solutions for the Networked Battlespace
Soraghan, J. & Weiss, S.
EPSRC (Engineering and Physical Sciences Research Council)
1/04/13 → 31/03/18
Project: Research

Cite this

@conference{1a67f4851f9241fd83fdecd3dcfe348d,

title = "Complexity and search space reduction in cyclic-by-row PEVD algorithms",

abstract = "In recent years, several algorithms for the iterative calculation of a polynomial matrix eigenvalue decomposition (PEVD) have been introduced. The PEVD is a generalisation of the ordinary EVD and uses paraunitary operations to diagonalise a parahermitian matrix. This paper addresses potential computational savings that can be applied to existing cyclic-by-row approaches for the PEVD. These savings are found during the search and rotation stages, and do not significantly impact on algorithm accuracy. We demonstrate that with the proposed techniques, computations can be significantly reduced. The benefits of this are important for a number of broadband multichannel problems.",

keywords = "polynomial matrix eigenvalue decomposition, PEVD, space-time covariance matrices, algorithmic cost reductions, SMDCbR, sequential matrix diagonalisation",

author = "Coutts, {Fraser K.} and Jamie Corr and Keith Thompson and Stephan Weiss and Proudler, {Ian K.} and McWhirter, {John G.}",

year = "2016",

month = jul,

day = "19",

language = "English",

note = "50th Asilomar Conference on Signals, Systems and Computers, Asilomar 2016 ; Conference date: 06-11-2016 Through 09-11-2016",

url = "http://www.asilomarsscconf.org/",

}

TY - CONF

T1 - Complexity and search space reduction in cyclic-by-row PEVD algorithms

AU - Coutts, Fraser K.

AU - Corr, Jamie

AU - Thompson, Keith

AU - Weiss, Stephan

AU - Proudler, Ian K.

AU - McWhirter, John G.

PY - 2016/7/19

Y1 - 2016/7/19

N2 - In recent years, several algorithms for the iterative calculation of a polynomial matrix eigenvalue decomposition (PEVD) have been introduced. The PEVD is a generalisation of the ordinary EVD and uses paraunitary operations to diagonalise a parahermitian matrix. This paper addresses potential computational savings that can be applied to existing cyclic-by-row approaches for the PEVD. These savings are found during the search and rotation stages, and do not significantly impact on algorithm accuracy. We demonstrate that with the proposed techniques, computations can be significantly reduced. The benefits of this are important for a number of broadband multichannel problems.

AB - In recent years, several algorithms for the iterative calculation of a polynomial matrix eigenvalue decomposition (PEVD) have been introduced. The PEVD is a generalisation of the ordinary EVD and uses paraunitary operations to diagonalise a parahermitian matrix. This paper addresses potential computational savings that can be applied to existing cyclic-by-row approaches for the PEVD. These savings are found during the search and rotation stages, and do not significantly impact on algorithm accuracy. We demonstrate that with the proposed techniques, computations can be significantly reduced. The benefits of this are important for a number of broadband multichannel problems.

KW - polynomial matrix eigenvalue decomposition

KW - PEVD

KW - space-time covariance matrices

KW - algorithmic cost reductions

KW - SMDCbR

KW - sequential matrix diagonalisation

UR - http://www.asilomarsscconf.org/

M3 - Paper

T2 - 50th Asilomar Conference on Signals, Systems and Computers

Y2 - 6 November 2016 through 9 November 2016

ER -

Complexity and search space reduction in cyclic-by-row PEVD algorithms

Abstract

Conference

Keywords

Access to Document

Fingerprint

Projects

Signal Processing Solutions for the Networked Battlespace

Cite this