| Title | Efficient Block and Time-Recursive Estimation of Sparse Volterra Systems |
| Authors | Stefan Ingi Adalbjörnsson, George-Othan Glentis, Andreas Jakobsson |
| Alternative Location | http://dx.doi.org/10.1109/S..., Restricted Access |
| Publication | 2012 IEEE Statistical Signal Processing Workshop (SSP), Proceedings of |
| Year | 2012 |
| Pages | 173 - 176 |
| Document type | Conference paper |
| Conference name | 2012 IEEE Statistical Signal Processing Workshop (SSP) |
| Conference Date | 2012-08-05/2012-08-08 |
| Conference Location | Ann Arbor, Michigan, USA |
| Status | Published |
| Quality controlled | Yes |
| Language | eng |
| Publisher | IEEE |
| Abstract English | We investigate the application of non-convex penalized least<br> squares for parameter estimation in the Volterra model. Sparsity<br> is promoted by introducing a weighted !q penalty on the<br> parameters and efficient batch and time recursive algorithms<br> are devised based on the cyclic coordinate descent approach.<br> Numerical examples illustrate the improved performance of<br> the proposed algorithms as compared the weighted !1 norm. |
| ISBN/ISSN/Other | ISBN: 978-1-4673-0183-1 (online) |
Questions: webmaster
Last update: 2013-04-11
Centre for Mathematical Sciences, Box 118, SE-22100, Lund. Telefon: +46 46-222 00 00 (vx)