優ガウス性の分布に基づくシステム同定のためのノンパラメトリックベイズモデル

カテゴリ: 論文誌(論文単位)

グループ名: 【C】電子・情報・システム部門

発行日: 2019/04/01

タイトル(英語): A Nonparametric Bayesian Model for System Identification Based on a Super-Gaussian Distribution

著者名: 丹治　寛樹（明治大学理工学部電気電子生命学科／日本学術振興会），村上　隆啓（明治大学理工学部電気電子生命学科），鎌田　弘之（明治大学理工学部電気電子生命学科）

著者名(英語): Hiroki Tanji (Department of Electronics and Bioinformatics, School of Science and Technology, Meiji University/JSPS), Takahiro Murakami (Department of Electronics and Bioinformatics, School of Science and Technology, Meiji University), Hiroyuki Kamata (Department of Electronics and Bioinformatics, School of Science and Technology, Meiji University)

キーワード: システム同定，優ガウス分布，majorization-minimizationアルゴリズム，ノンパラメトリックベイズモデル，beta-Bernoulli過程，ギブスサンプラー　　system identification，super-Gaussian distribution，majorization-minimization algorithm，nonparametric Bayesian model，beta-Bernoulli process，Gibbs sampler

要約(英語): In the acoustic signal processing applications of finite impulse response (FIR) system identification, it is important to develop the identification method that is robust to super-Gaussian noises. Moreover, the identification method that estimates the FIR coefficients and the order of the unknown system is required, because the order of the unknown system is unavailable in advance. Therefore, in this paper, we propose a nonparametric Bayesian (NPB) model for FIR system identification using a super-Gaussian likelihood and the beta-Bernoulli process. In the proposed NPB model, we employ the hyperbolic secant distribution for the likelihood function. Then, we derive the inference algorithm to simultaneously estimate the FIR coefficients and the order of the unknown system. Our inference algorithm based on a hybrid inference approach combining the majorization-minimization (MM) algorithm and the Gibbs sampler. The simulation results suggest that the proposed method outperforms the conventional identification algorithms in a super-Gaussian noise environment.

本誌: 電気学会論文誌C（電子・情報・システム部門誌） Vol.139 No.4 （2019） 特集：スマートシステムと計測・制御技術

本誌掲載ページ: 380-387 p

原稿種別: 論文／日本語

電子版へのリンク: https://www.jstage.jst.go.jp/article/ieejeiss/139/4/139_380/_article/-char/ja/