講演抄録/キーワード |
講演名 |
2009-06-11 10:25
確率的最大尤度方位推定法の厳密な定式化 ○鈴木正清・陳 海華(北見工大) SIS2009-2 |
抄録 |
(和) |
本稿では,確率的あるいは条件なし最大尤度(Stochastic or Unconditional
Maximum Likelihood, UML)方位推定法の厳密な定式化を提案する.これまでの定式化では,重要な条件,すなわちノイズ成分を除く受信信号成分の共分散行列の非負定値性が抜け落ちており,かつ,不適切な大局解が解空間に現れる.本稿で提案する厳密な定式化によって,これらの問題が解決される. |
(英) |
This paper presents an exact formulation of Stochastic or
Unconditional Maximum Likelihood (UML) estimation for directions-of-arrival (DOA) finding. In the previous formulation, an important condition is missing. That is the non-negative definiteness of the covariance matrix of signal components without additive noises. Because of this, it is not only guaranteed that the covariance matrix is non-negative definite, but also inadequate global solutions appear in the solution space. Furthermore global search fails in finding an adequate solution. This paper proposes an exact formulation of UML which guarantees to estimete a non-negative definite covariance matrix of signal components. Then the inadequate global solutions disappear and global search succeeds in finding adequate solution. |
キーワード |
(和) |
方位推定 / 最尤法 / 確率信号モデル / 非負定値性 / / / / |
(英) |
DOA finding / ML estimation / tochastic signal model / non-negative definiteness / / / / |
文献情報 |
信学技報, vol. 109, no. 78, SIS2009-2, pp. 7-12, 2009年6月. |
資料番号 |
SIS2009-2 |
発行日 |
2009-06-04 (SIS) |
ISSN |
Print edition: ISSN 0913-5685 Online edition: ISSN 2432-6380 |
著作権に ついて |
技術研究報告に掲載された論文の著作権は電子情報通信学会に帰属します.(許諾番号:10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) |
PDFダウンロード |
SIS2009-2 |