For　the　problem　of　non-stationary　signals　of　rolling　bearing　and　the　difficulty　to　get　a　large　number　of　typical　fault　samples　in　practice,　a　fault　diagnosis　method　was　proposed　based　on　dual-tree　complex　wavelet　packet　transform　and　support　vector　machine　(SVM).　Firstly,　the　non-stationary　fault　vibration　signal　was　decomposed　into　several　different　frequency　band　components　through　dual-tree　complex　wavelet　packet　transform;　secondly,　normalization　processing　was　made　from　the　energy　of　each　component.　Finally,　the　energy　characteristics　parameters　of　each　frequency　band　component　were　taken　as　input　of　the　SVM　to　identify　the　fault　type　of　rolling　bearing.　The　analog　signals　of　experiments,　containing　normal　condition　of　the　rolling　bearing,　crack　fault　of　bearing　outer　ring,　crack　fault　of　bearing　inner　ring　and　pitting　fault　of　bearing　rolling　element,　were　analyzed　and　the　fault　recognition　rate　reaches　99.　5%.　The　proposed　method　can　identify　the　working　state　and　fault　type　of　rolling　bearing　accurately　and　effectively,　as　compared　with　the　method　of　combining　traditional　wavelet　packet　transform　with　SVM.