摘要 |
Disclosed are an apparatus and a method for eliminating noise from a biomedical signal. The apparatus for eliminating noise from a biomedical signal includes a wavelet learning unit for generating a probability table showing frequency of selecting a base wavelet function as probabilities through a wavelet train based on an input of a non-noise biomedical signal, and a noise eliminating unit for receiving a biomedical signal including noise and for performing wavelet conversion and wavelet reduction by using an optimal wavelet function selected based on the probability table to eliminate the noise from the biomedical signal. According to the present invention, noise generated in various experiment environments such as EEG (electroencephalography), EMG (electromyography), ECG (Electrocardiogram) or FNIRS (Functional Near Infrared Spectroscopy) may be optimally reduced. Furthermore, since the noise elimination is required in BSI and biomedical signal based diagnosis supplementary software technique, there is a strong possibility of commercializing the apparatus according to the present invention in future. |