Nonlinear system identification using neural networks trained with natural gradient descent (Q1886938)
From MaRDI portal
 | This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: Nonlinear system identification using neural networks trained with natural gradient descent |
scientific article; zbMATH DE number 2117014
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Nonlinear system identification using neural networks trained with natural gradient descent |
scientific article; zbMATH DE number 2117014 |
Statements
Nonlinear system identification using neural networks trained with natural gradient descent (English)
0 references
19 November 2004
0 references
An unknown nonlinear system is assumed to be a blackbox and the learning process is performed using the input-output signal only. Natural gradient (NG) learning neural networks (NNs) are used for modeling nonlinear systems with memory. The NN model is composed of a linear adaptive filter followed by a two-layer memoryless nonlinear NN. A Kalman-filter-based technique and a search-and-converge method are employed for the NG algorithm. Computer simulations and illustrations are given.
0 references
satellite communication
0 references
system identification
0 references
adaptive signal processing
0 references
neural networks
0 references
natural gradient learning
0 references
nonlinear systems with memory
0 references
Kalman filter
0 references
