Presentation | 2000/3/16 Robust Identification of Nonlinear System by Reduced Rank Volterra Filter Tomohiro Sekiguchi, Isao Yamada, Kohichi Sakaniwa, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | In this paper, we propose the reauced rank polynomial filter (RRPE), a vector valued polynomial operator whose dominant kernel (coefficient matrix) has its rank restricted arbitrarily. It is shown that a drastic reduction of computational complexity is achieved by the rank reduction of the dominant kernel. More importantly, the rank reduction can make the polynomial filter more robust to noise by restricting the influences of higher order multiplications of noise within the reduced column space of the dominant kernel. The main theorem of the paper presents an explicit formula to determine recursively all the coefficient matrices of the optimal RRPF named the reduced rank Volterra filter (RRVF) that minimizes the mean square error (MSE) between its output vector eandom variable and the desired vector random variable to be approximated. A numerical example indicates that the proposed reduced rank Volterra filter (RRVF) realizes an ideal trade-off between the computational complexity and the estimation accuracy. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Volterra filter / Reduced rank Volterra filter / Nonlinear filter / Higher order statistics / Robust estimation |
Paper # | CAS99-122,DSP99-192,CS99-163 |
Date of Issue |
Conference Information | |
Committee | DSP |
---|---|
Conference Date | 2000/3/16(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 | Digital Signal Processing (DSP) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Robust Identification of Nonlinear System by Reduced Rank Volterra Filter |
Sub Title (in English) | |
Keyword(1) | Volterra filter |
Keyword(2) | Reduced rank Volterra filter |
Keyword(3) | Nonlinear filter |
Keyword(4) | Higher order statistics |
Keyword(5) | Robust estimation |
1st Author's Name | Tomohiro Sekiguchi |
1st Author's Affiliation | Department of Electrical and Electronic Engineering Tokyo Institute of Technology() |
2nd Author's Name | Isao Yamada |
2nd Author's Affiliation | Department of Electrical and Electronic Engineering Tokyo Institute of Technology |
3rd Author's Name | Kohichi Sakaniwa |
3rd Author's Affiliation | Department of Electrical and Electronic Engineering Tokyo Institute of Technology |
Date | 2000/3/16 |
Paper # | CAS99-122,DSP99-192,CS99-163 |
Volume (vol) | vol.99 |
Number (no) | 695 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |