De-noising of EEG signal for emotion recognition

Brain Computer Interface (BCI) systems are the devices which are proposed to help the patients incapable of making motor response to communicate with external device like computer using brain signal. Signal pre-processing is the first phase in emotion recognition from EEG signal. This part of BCI is the most vital part of the system as the accuracy of the system determined in the next phase will be based on the acquired results. Hence, several different techniques have been reported that quantify the information of the brain signal. In this paper, the performance of DWT db4 level 1 has been compared with other existing algorithms in terms of SNR, PSNR and Noise Covariance for de-noising the signal.