Electroencephalogram (EEG) is the recording of electrical activity along the scalp. EEG is very small signals of a few μV. Artifacts arising from living body often mingle with the EEG. A brain computer interface (BCI) is a communication between a human brain and an external device using EEG. The artifacts interfere with driving the BCI system. They may change the features of nervous activity or even be incorrectly used as the source of control in BCI systems. Recently, independent component analysis (ICA) has become one of the most prominent techniques for EEG. A typical method removes the artifacts after the offline training using the ICA. However, the method is not suitable for BCI systems such as online driving. So, we propose a method to eliminate artifacts (especially blink artifacts) from the body. This method does not require offline training and processes close to the real-time processing.