Presentation 2008-06-27
Avoidance of Numerical Unstability in Exact Unconditional ML Estimation of DOA
Haihua CHEN, Masakiyo SUZUKI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper addresses the issue of numerical unstability in Unconditional Maximum Likelihood (UML) estimation of Directions-of-Arrival (DOA) finding. First we show brief description of the definition of the exact UML estimation and the conventional method solving it, Alternating Minimization (AM) method. The exact UML criterion is formulated by using orthonormal bases of the signal and noise subspaces. When more than one directional parameters have an identical value, the dimension of the signal subspace decreases. In other words, in the parameter space the change of each subspace is discontinuous at the points that more than one parameters have the identical value. As a result, when more than one directional parameters converge to an identical value, the calculation of the bases of these two subspaces becomes numerically unstable. In this paper, we propose a technique which eliminates the discontinuance with respect to the change of the noise subspace to make the UML estimation numerically stable. Finally, simulation results are also shown to demonstrate the validity of the proposed technique.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) DOA finding / Stochastic signal model / ML / Numerical unstability / Discontinuance of subspace
Paper # CAS2008-29,VLD2008-42,SIP2008-63
Date of Issue

Conference Information
Committee VLD
Conference Date 2008/6/20(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To VLSI Design Technologies (VLD)
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Avoidance of Numerical Unstability in Exact Unconditional ML Estimation of DOA
Sub Title (in English)
Keyword(1) DOA finding
Keyword(2) Stochastic signal model
Keyword(3) ML
Keyword(4) Numerical unstability
Keyword(5) Discontinuance of subspace
1st Author's Name Haihua CHEN
1st Author's Affiliation Graduate School of Engineering, Kitami Institute of Technology()
2nd Author's Name Masakiyo SUZUKI
2nd Author's Affiliation Graduate School of Engineering, Kitami Institute of Technology
Date 2008-06-27
Paper # CAS2008-29,VLD2008-42,SIP2008-63
Volume (vol) vol.108
Number (no) 107
Page pp.pp.-
#Pages 6
Date of Issue