DNA　microarrays　can　be　used　to　measure　the　expression　levels　of　thousands　of　genes　simultaneously　and　so　it　is　usually　an　effective　tool　of　cancer　research.　A　discrete　wavelet　transform(DWT)　based　feature　extraction　method　for　tissue　classification　was　introduced,　by　which　microarray　expression　data　are　transformed　into　time-scale　domain　and　used　as　input　of　a　tissue　classifier.　Finally,　some　test　and　comparison　experiments　for　the　feature　extraction　method　have　been　made　by　using　the　weighted　voting　classification　scheme　[1].　Experiment　results　show　that　the　correct　rate　is　at　least　90%　in　tumor　vs.　normal　classification　by　using　the　feature　extraction　method.