A nonparametric Bayesian model for system identification based on a super-Gaussian distribution

被引：0
作者：
Tanji H. [1 ]
Murakami T. [2 ]
Kamata H. [2 ]
机构：
[1] Department of Electronics and Bioinformatics, School of Science and Technology, Meiji University JSPS Research Fellowships for Young Scientists (DC2), 1-1-1, Higashi-mita, Tama-ku, Kawasaki, Kanagawa
[2] Department of Electronics and Bioinformatics, School of Science and Technology, Meiji University, 1-1-1, Higashi-mita, Tama-ku, Kawasaki, Kanagawa
来源：
IEEJ Transactions on Electronics, Information and Systems | 2019年 / 139卷 / 04期
基金：
日本学术振兴会;
关键词：
Beta-Bernoulli process; Gibbs sampler; Majorization-minimization algorithm; Nonparametric Bayesian model; Super-Gaussian distribution; System identification;
D O I：
10.1541/ieejeiss.139.380
中图分类号：
学科分类号：
摘要：
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. © 2019 The Institute of Electrical Engineers of Japan.
引用
收藏
页码：380 / 387
页数：7
相关论文
共 29 条
[1]  
Isermann R., Munchhof M., Identification of Dynamic Systems, (2011)
[2]  
Schmidt G., Applications of acoustic echo control-An overview, Proc. of 12th European Signal Processing Conference (EUSIPCO), pp. 9-16, (2004)
[3]  
Kuo S.M., Morgan D.R., Active noise control: A tutorial review, Proceedings of the IEEE, 87, 6, pp. 943-973, (1999)
[4]  
Haykin S., Adaptive Filter Theory, (2002)
[5]  
Bottegal G., Aravkin A.Y., Hjalmarsson H., Pillonetto G., Robust em kernel-based methods for linear system identification, Automatica, 67, pp. 114-126, (2016)
[6]  
Prando G., Chiuso A., Pillonetto G., Maximum entropy vector kernels for MIMO system identification, Automatica, 79, pp. 326-339, (2017)
[7]  
Bottegal G., Hjalmarsson H., Aravkin A.Y., Pillonetto G., Outlier robust kernel-based system identification using l1-Laplace techniques, Proc. of 54th IEEE Conference on Decision and Control (CDC), pp. 2109-2114, (2015)
[8]  
Tanji H., Tanaka R., Murakami T., Ishida Y., FIR system identification based on a nonparametric Bayesian model using the Indian buffet process, Signal, Image, and Video Processing, 10, 6, pp. 1105-1112, (2016)
[9]  
Rasmussen C.E., Williams C.K.I., Gaussian Processes for Machine Learning, (2006)
[10]  
Gazor S., Zhang W., Speech probability distribution, IEEE Signal Processing Letters, 10, 7, pp. 204-207, (2003)
123