In light of prediction of remaining useful life (RUL) of rolling bearings, a long short-term memory (LSTM) network is introduced into traditional methods. First, the feature of rolling bearings is extracted from the time domain, the frequency domain and the time-frequency domain. Then, three evaluation indexes are defined to characterize the degeneration, and the data is filtered for a degenerated feature set to train the LSTM network prediction model. Finally, the remaining useful life is predicted by the trained neural network. The proposed method is accurate in prediction and superior to back propagation (BP) neural network and support vector regression