Abstract: A similarity measure about n -dimensional Boolean pattern is proposed in this paper. With this similarity measure, two kinds of clustering algorithms are constructed. These algorithms are simple, parallel, and can be easily implemented with hardware, but they are sensitive to variance.