Services on Demand
Journal
Article
Indicators
- Cited by SciELO
- Access statistics
Related links
- Cited by Google
- Similars in SciELO
- Similars in Google
Share
Revista de Ingeniería
Print version ISSN 0121-4993
Abstract
PADILLA BURITICA, Jorge Iván; AVENDANO VALENCIA, Luis David and CASTELLANOS DOMINGUEZ, Germán. Improved Estimation of Multivariate AR Model for EEG Signal Processing. rev.ing. [online]. 2013, n.38, pp.20-26. ISSN 0121-4993.
This paper proposes a methodology to enhance the estimation of parameter matrices in multivariate autoregressive models. It uses the Kalman filter and state space representation to improve the precision of parameter estimation, while maintaining a reduced computational burden. Two methods of covariance matrix adaptation are considered within the Kalman filter framework to improve the estimator convergence rate while preserving precision. The methodology is tested on simulated data, electroencephalographic recordings and performance. As a result, a reduction of the computational burden of up to 40% is achieved, but reconstruction error reaches as much as 3%
Keywords : Electroencephalography; multivariate autoregressive models; Kalman filter; forgetting factor; simulated annealing.