<p>Recently wearable electroencephalography(EEG) device enables us to mea sure brain activities during moving. However, EEG data include noise caused by movement when participants move. Therefore, it is a significant problem to reduce the noise in order to measure brain activities during moving. In this paper, we focused on the workload estimation from auditory steady state response during walking. We compared the noise reduction performance of some kinds of blind source separation methods.</p>