Monitoring of incipient faults is vital for the safe and reliable operation of rolling element bearings. In this paper, the combination of discrete wavelet transform and envelope analysis is proposed to extract the characteristic spectrum of rolling bearing vibration data. Then spectrum cross-correlation coefficient is applied to identify different operating conditions of rolling bearings. The proposed method is applied to fault diagnosis of rolling bearings with several different faults. The results show that the proposed method has high classification accuracy, and performs better than alternative approaches based on conventional characteristic defect frequency extraction.