Abstract
A method is introduced for estimating unknown signals corrupted by a noise that can be modeled by an autoregressive (AR) process. The filter, referred to as the predictor-subtractor-restorer (PSR) filter is used to estimate average visual evoked responses that are corrupted by the background EEG. It is demonstrated that this filter offers superior performance to previously reported Wiener filters that have been used in this application.
Original language | English (US) |
---|---|
Pages (from-to) | 414-417 |
Number of pages | 4 |
Journal | Unknown Journal |
Volume | 8 |
Issue number | 5 |
DOIs | |
State | Published - 1978 |